prospects for co2 capture and storage - ccs-info.org · 2014-07-07 · 8 prospects for co2 capture...

INTERNATIONAL ENERGY AGENCY

EnergyTechnologyAnalysis

PROSPECTSFOR

CO2 CAPTUREAND

STORAGE

INTERNATIONAL ENERGY AGENCY9, rue de la Fédération,

75739 Paris Cedex 15, France

The International Energy Agency (IEA) is anautonomous body which was established inNovember 1974 within the framework of theOrganisation for Economic Co-operation andDevelopment (OECD) to implement an inter-national energy programme.

It carries out a comprehensive programme ofenergy co-operation among twenty-six* of theOECD’s thirty Member countries.The basic aimsof the IEA are:

• to maintain and improve systems for copingwith oil supply disruptions;

• to promote rational energy policies in a globalcontext through co-operative relations withnon-member countries, industry and inter-national organisations;

• to operate a permanent information system onthe international oil market;

• to improve the world’s energy supply anddemand structure by developing alternativeenergy sources and increasing the efficiency ofenergy use;

• to assist in the integration of environmental andenergy policies.

* IEA Member countries: Australia, Austria, Belgium,Canada, the Czech Republic, Denmark, Finland,France, Germany, Greece, Hungary, Ireland, Italy,Japan, the Republic of Korea, Luxembourg,the Netherlands, New Zealand, Norway, Portugal,Spain, Sweden, Switzerland, Turkey, the UnitedKingdom, the United States. The EuropeanCommission also takes part in the work of the IEA.

ORGANISATION FORECONOMIC CO-OPERATION

AND DEVELOPMENT

Pursuant to Article 1 of the Convention signed inParis on 14th December 1960, and which cameinto force on 30th September 1961, the Organisationfor Economic Co-operation and Development(OECD) shall promote policies designed:

• to achieve the highest sustainable economicgrowth and employment and a rising standardof living in Member countries, while maintainingfinancial stability, and thus to contribute to thedevelopment of the world economy;

• to contribute to sound economic expansion inMember as well as non-member countries in theprocess of economic development; and

• to contribute to the expansion of world tradeon a multilateral, non-discriminatory basis inaccordance with international obligations.

The original Member countries of the OECD areAustria, Belgium, Canada, Denmark, France,Germany, Greece, Iceland, Ireland, Italy,Luxembourg, the Netherlands, Norway, Portugal,Spain, Sweden, Switzerland, Turkey, the UnitedKingdom and the United States. The followingcountries became Members subsequently throughaccession at the dates indicated hereafter: Japan(28th April 1964), Finland (28th January 1969),Australia (7th June 1971), New Zealand (29th May1973),Mexico (18th May 1994), the Czech Republic(21st December 1995), Hungary (7th May 1996),Poland (22nd November 1996), the Republic ofKorea (12th December 1996) and Slovakia(28th September 2000). The Commission of theEuropean Communities takes part in the work ofthe OECD (Article 13 of the OECD Convention).

© OECD/IEA, 2004

Applications for permission to reproduce or translate all or part of this publication should be made to:Head of Publications Service, OECD/IEA

2, rue André-Pascal, 75775 Paris Cedex 16, Franceor

9, rue de la Fédération, 75739 Paris Cedex 15, France.

001-006 17/11/2004 12:00

FOREWORD

Fossil fuels will be used extensively and CO2 emissions will rise over the next half century, if nonew policies are put in place. It is clear that such a development is not sustainable. A number ofoptions exist that can reduce the CO2 emissions from the energy system. These include improvedenergy efficiency and a switch to renewable and nuclear energy. However, policies based on theseoptions will, at best, only partly solve the problem. Carbon dioxide capture and storage (CCS)technologies constitute another promising option that can drastically reduce these emissions. Toaccomplish this, governments need to take action now to ensure that CCS technologies are developedand deployed on a large scale over the next few decades.

This publication describes the challenges that must be overcome for a CCS strategy to reachmarket introduction and achieve its full potential within the next 30-50 years. The quantitativeand qualitative analyses in this book reveal that large-scale uptake of capture and storagetechnologies is probably 10 years off, and that without a major increase in RD&D investment, thetechnology will not be in place to realize its full potential in the coming decades. Effective emissionreduction incentives will be needed to achieve market deployment from 2015 onward. Additionalpolicies will have to be implemented to remove barriers and reduce uncertainties. If the rightaction is taken, CCS could become an essential ‘transition technology’ to a sustainable energysystem for the next 50 to 100 years.

This volume shows how CCS technologies can help to reduce emissions significantly over the comingtwo to five decades. The analysis includes three important elements. First, it provides a comprehensiveoverview of the prospects, costs and RD&D challenges of CO2 capture, transportation and storagetechnologies. Second, using a newly developed energy technology model, it presents global scenarioanalyses that investigate how CCS technologies can contribute to reducing CO2 emissions and whatconditions would be necessary to justify stepping up RD&D and international efforts to advanceCCS. Finally, the third element highlights the priority policy actions that should be taken to ensurethe timely deployment of CCS technologies.

The results suggest that an aggressive policy of developing and deploying CCS technologies couldindeed achieve substantial reductions in worldwide CO2 emissions. Although the main role of CCSwould be in the electricity sector, interesting possibilities also exist in manufacturing and in theproduction of transportation fuels. With sufficient technology investment, and after successfullysolving various environmental and legal issues, CCS can provide a way to curb greenhouse gasemissions substantially at acceptable costs in most areas in the world.

This publication is the first in a new IEA series entitled Energy Technology Analysis. The goal ofthis series is to use quantitative model analysis for the assessment of the prospects of emergingenergy technologies and their potential impact on energy supply security, economic developmentand the environment. I am confident that this analysis provides new insights for IEA membergovernments and other decision makers on how to use innovative technologies as part of an efficientlong-term emissions mitigation strategy.

Claude MandilExecutive Director

FOREWORD 3

001-006 17/11/2004 12:00

001-006 17/11/2004 12:00

ACKNOWLEDGEMENTS

The main authors of this book are Dolf Gielen and Jacek Podkanski, with contributions fromFridtjof Unander. The work was done under the supervision of Marianne Haug, Director of the EnergyTechnology and R&D Office of the International Energy Agency. It is a result of close cooperationbetween the Energy Technology Policy Division, headed by Carmen Difiglio until September 2004and by Fridtjof Unander thereafter, and the Energy Technology Collaboration Division, headed byAntonio Pflüger. The study also benefited from input by other IEA colleagues, namely: Fatih Birol,Rick Bradley, John Cameron, François Cattier, Niclas Mattsson, Cédric Philibert and Rick Sellers.

A number of experts provided input to the study, commented on the underlying analytical work,reviewed early drafts of the manuscript and provided guidance for the model development. Theircomments and suggestions were of great value. Prominent contributors include: Stefan Bachu(Alberta Geological Survey, Canada), Graham Campbell (Natural Resources Canada), Carmen Difiglio(Department of Energy, USA), John Davison (IEA Greenhouse Gas R&D Programme, UK), AugustinFlory (Consultant, France), Paul Freund (IEA Greenhouse Gas R&D Programme, UK), John Gale(IEA Greenhouse Gas and R&D Programme, UK), Gary Goldstein (International Resources Group,USA), Ian Hayhow (Natural Resources Canada), Hubert Höwener (Research Center Jülich, Germany),Olav Kaarstad (Statoil ASA, Norway), Jostein Dahl Karlsen (Ministry of Petroleum and Energy,Norway), John “Skip” Laitner (EPA, USA), Barbara McKee (Department of Energy, USA), PeterMarkewitz (Research Center Jülich, Germany), Hanns-Joachim Neef (Research Center Jülich, Germany),Ken Noble (Noble-Soft, Australia), Ed Rubin (Carnegie Mellon University, USA), Michael Sachse(Research Center Jülich, Germany), Jochen Seier (Research Center Jülich, Germany), Koen Smekens(Energy research Centre of the Netherlands), John Topper (IEA Clean Coal Centre, UK), GianCarloTosato (ETSAP and Max-Planck-Institut für Plasmaphysik, Germany), Clas-Otto Wene (Consultant,Sweden), the participants of the IEA Energy Technology Systems Analysis Programme (ETSAP) andthe members of the IEA Working Party on Fossil Fuels. All errors and omissions are solely theresponsibility of the authors.

The Energy Technology Perspectives model development could not have been achieved withoutsubstantial financial support provided by the governments of many countries, notably Canada,Japan, Norway, Sweden, the United Kingdom and the United States.

The manuscript was edited by Sally Bogle, Kathleen Gray and Geoff Morrison. Assistance withpreparation of the manuscript was provided by Muriel Custodio, Corinne Hayworth and LorettaRavera.

ACKNOWLEDGEMENTS 5

001-006 17/11/2004 12:00

001-006 17/11/2004 12:00

TABLE OF CONTENTS

Foreword 3

Acknowledgements 5

Overview 13

1. Introduction 23

The purpose of the study 24

The structure of the study 24

2. The world energy system, CO2 emissions and mitigationoptions 27

Global CO2 emissions: past trends and future outlook 27

Factors affecting and strategies to reduce CO2 emissions 29

3. CCS characteristics: technologies, potential, permanence and cost 37

General characteristics of CO2 capture and storage 40

CO2 capture opportunities in the world energy system 42

CO2 capture in the electricity sector 48

CO2 capture options for power plants 48Efficiency first – clean coal technologies 59

CO2 capture in the manufacturing industry 66

Process industries 66Furnaces and CHP 68

CO2 capture in fuels supply 69

Natural gas processing 70Oil refineries 70Hydrogen production 74Gasification and Fischer-Tropsch production of liquid synfuels 75

Technology learning effects for CO2 capture 78

CO2 transportation 79

CO2 storage 81

Depleted oil and gas fields 83CO2 enhanced oil recovery 84CO2 enhanced gas recovery 87CO2 enhanced coal-bed methane recovery 88Storage in deep saline aquifers 90Other storage options 93

TABLE OF CONTENTS 7

007-012 18/11/2004 12:24

8 PROSPECTS FOR CO2 CAPTURE AND STORAGE

CO2 storage: permanence and monitoring 94

Production of chemicals and fuels from CO2 97

Overview of CCS costs 98

4. Basic results from the model analysis 101

The Energy Technology Perspectives (ETP) Model 102

Overview of the model analysis structure 103

The BASE scenario 105

The GLO50 scenario 106

CO2 capture in the electricity sector 112CO2 capture in the manufacturing industry 116CO2 capture in the fuels supply 117CO2 storage 118

CCS compared to other emission reduction options 119

5. CCS sensitivity analysis 125

CO2 policy targets 127

CO2 policy scope and timing 129

GDP growth and energy demand 131

Renewables 133

Nuclear power 138

CCS technology progress 139

Market structure 141

Fuel prices 141

Analysis time horizon 143

Overview of sensitivity analysis results 143

6. Regional activities and CCS scenario analysis 145

Global CCS scenario analysis 146

CCS potentials and RD&D activities in a regional perspective 153

North America 154

Europe 158

Asia-Pacific OECD countries 161

China 163

India 164

Middle East 165

007-012 18/11/2004 12:24

7. The impact of CCS on energy markets: model results 167

Coal 168

Gas 171

Oil 173

Renewables 175

Electricity 178

8. Challenges ahead and priorities for action 181

Interrelated challenges 182

Timing issues 183

RD&D challenges 184

Public awareness and acceptance 189

The regulatory and legal framework 190

Long-term policy framework and incentives 197

Annexes 201

Annex 1 ETP model caracteristics 201

Annex 2 Regional investment costs and discount rates 219

Annex 3 GDP projections 221

Annex 4 Websites for more information on CCS 225

Annex 5 Definitions, abbrevations, acronyms and units 227

References 235

List of figures

Figure 2.1 Energy-related CO2 emissions, globally and by region (1973-2030)Figure 2.2 Energy-related CO2 emissions by fuel (1973-2030)Figure 2.3 CO2 emissions in the IPCC SRES A1 scenarios, compared to the WEO Reference

ScenarioFigure 2.4 Growth in GDP and CO2 emissions: decomposition of factors affecting the linkFigure 2.5 CO2 intensity in global electricity generation and fossil fuelsFigure 2.6 Growth in electricity demand and CO2 emissions from power generation:

decomposition of factors affecting the linkFigure 2.7 Share of fuels in global electricity generationFigure 2.8 Reduction in energy-related CO2 emissions in the WEO Alternative Policy Scenario,

by contributory factor

TABLE OF CONTENTS 9

007-012 18/11/2004 12:24


Figure 3.1 Investment capital cost shares for oxyfuel retrofit of coal-fired power plants Figure 3.2 European coal-fired power plant building activity Figure 3.3 North American coal-fired power plant building activity Figure 3.4 Japanese coal-fired power plant building activity Figure 3.5 Chinese coal-fired power plant building activity Figure 3.6 Power plant efficiencies as a function of the cycle temperature Figure 3.7 CO2 emissions from oil refining Figure 3.8 Investment cost structure for a refinery complex with CO2 capture Figure 3.9 Hydrogen production costs for a fully developed supply system Figure 3.10 Cost of overland transportation of CO2 by pipeline Figure 3.11 Onshore oil well cost as a function of the well depth Figure 3.12 Known conventional petroleum reserves of the world by region Figure 3.13 Cost breakdown of a proposed ECBM project in the Netherlands Figure 3.14 The cost structure of Norway’s Snohvit aquifer storage project Figure 4.1 The role of ETP model runs in the analysisFigure 4.2 Primary energy demand projections in the BASE scenarioFigure 4.3 GHG penalties in the GLO50 scenarioFigure 4.4 Global CO2 emissions, BASE and GLO50 scenarios Figure 4.5 Primary energy mix in the GLO50 scenarioFigure 4.6 Global CO2 capture by process area, GLO50 scenarioFigure 4.7 Total CO2 capture split by technology, GLO50 scenarioFigure 4.8 The electricity production mix, GLO50 scenario Figure 4.9 Efficiency trends for coal and gas-fired power plantsFigure 4.10 Electricity production from power plants fitted with CCS, by technology and fuel,

GLO50 scenarioFigure 4.11 CO2 capture in the manufacturing industry, GLO50 scenarioFigure 4.12 CO2 capture in the fuels supply sector, GLO50 scenarioFigure 4.13 CO2 storage in the GLO50 scenarioFigure 4.14 CO2 emissions with and without CCSFigure 4.15 Cumulative emission abatement for 2000-2050 as a function of the penalty levelFigure 4.16 Marginal emission reduction cost with various sets of options availableFigure 4.17 Average cost of emission reduction with various sets of options availableFigure 5.1 CO2 capture at various policy incentive levelsFigure 5.2 Energy related and inorganic CO2 emissions at various policy incentive levels,

compared to long-term stabilization scenarios at 550 ppm and 450 ppm Figure 5.3 Share of CCS in total CO2 emissions mitigationFigure 5.4 CO2 capture for a 50 USD/t case with global policy targets and OECD policy targetsFigure 5.5 The impact of GDP growth on the global use of CCS Figure 5.6 Electricity production, GLO50nuclearFigure 5.7 CO2 capture by region, GLO50 and GLO50nuclearFigure 5.8 CO2 capture without IGCC transportation fuel cogenerationFigure 6.1 CO2 emissions in the four scenarios, 2000-2050Figure 6.2 CO2 capture in the four scenarios, 2000-2050Figure 6.3 CO2 capture in the four scenarios by technology, 2030 and 2050Figure 6.4 CO2 capture in the four scenarios by fuel type, 2030 and 2050Figure 6.5 Primary fuel mix in the four scenarios, 2030 and 2050Figure 6.6 Electricity production by fuel type, 2030 and 2050Figure 6.7 Electricity production by power plants fitted with CCS technology, by region

(2030 and 2050, GLO50 scenario)

007-012 18/11/2004 12:24

Figure 6.8 CO2 capture in North America in the four scenariosFigure 6.9 CO2 capture in Eastern and Western Europe in the four scenariosFigure 6.10 CO2 capture in the OECD Asia-Pacific region in the four scenariosFigure 6.11 CO2 capture in China in the four scenariosFigure 6.12 CO2 capture in India in the four scenariosFigure 6.13 CO2 capture in the Middle East in the four scenariosFigure 7.1 Coal use under various CO2 penalty levels, if CCS is considered Figure 7.2 Relative change in coal use without CCS vs. with CCSFigure 7.3 Gas use under various CO2 penalty levels, if CCS is consideredFigure 7.4 Crude and syncrude use under various CO2 penalty levels, if CCS is consideredFigure 7.5 Relative change in crude and syncrude use without CCS vs. with CCS Figure 7.6 Biomass use under various CO2 penalty levels, if CCS is considered Figure 7.7 Biomass use in the GLO50 scenario (with CCS)Figure 7.8 The use of other renewables under various CO2 penalty levels,

if CCS is consideredFigure 7.9 Relative change in other renewables use without CCS vs. with CCS Figure 8.1 IEA government RD&D budgetsFigure 8.2 Trajectories for increased efficiency and CCS developmentFigure 8.3 Roadmap for efficiency improvements and CCS development in coal-fired power

plants in OECD countriesFigure 8.4 Roadmap for efficiency improvements and CCS development in coal-fired power

plants in non-OECD countriesFigure 8.5 Major CO2 storage projects and the uncertain

long-term developments (cumulative)Figure 8.6 UNCLOS legal zones of the sea Figure A1.1 The ETP model reference energy systemFigure A1.2 Oil and gas production moduleFigure A1.3 Structure for gas supply in the ETP modelFigure A1.4 Structure for coal in the ETP modelFigure A1.5 Schematic CCS model structure for the electricity sector Figure A1.6 ETP wind electricity supply curve for EuropeFigure A1.7 ETP electricity load curve for Western Europe.

List of tables

Table 2.1 Impact on global CO2 emissions reductions from different factorsTable 3.1 Aggregate world energy balance by source category (2000)Table 3.2 Global CO2 emissions by source category (2000)Table 3.3 Commercial CO2 scrubbing solvents used in industry Table 3.4 Characteristics of power plants with and without CO2 captureTable 3.5 Average regional efficiencies for centralized, coal- and gas-fired power plants

(2000) Table 3.6 Characteristics of furnace/CHP unit with CO2 capture Table 3.7 Regional refinery structure (2000) Table 3.8 CO2 emissions in various refining and synfuel production processes Table 3.9 CO2 pressurization energy requirements for injection as a function of type of

reservoir and depth

TABLE OF CONTENTS 11

007-012 18/11/2004 12:24


Table 3.10 Characteristics of CO2 - enhanced fossil fuel production Table 3.11 CO2 EOR projects worldwide Table 3.12 Screening criteria for enhanced oil recovery methods Table 3.13 Recent estimates of CO2 storage potentials in deep saline aquifers Table 3.14 Overview of likely CCS costsTable 5.1 Overview of ETP sensitivity analysisTable 5.2 Renewable electricity deployment targets in the sensitivity analysis GLO50RENTable 5.3 Learning rates and investment costs used in the ETP modelTable 5.4 Electricity production by fuel and technology category for various learning

assumptions for renewables, 2050Table 5.5 Cost reductions for PV as a function of the target levelTable 5.6 Cost reductions for wind as a function of the target level Table 5.7 Change in CCS use in a liberalized market Table 5.8 Overview of sensitivity analysis results Table 6.1 Characteristics of the ETP model’s EFTEP scenariosTable 6.2 CSLF international co-peration projects Table 7.1 Model coal price changes under various CO2 penalty levels,

compared to BASE (2040)Table 7.2 Model natural gas price changes under various CO2 penalty levels,

compared to BASE (2040) Table 7.3 Model oil price changes under various CO2 penalty levels,

compared to BASE (2040) Table 7.4 Electricity production by fuel type, with and without CCS technology

(2030 and 2050)Table 7.5 Model electricity price increase under various CO2 penalty levels,

with and without CCS technology (2040)Table 8.1 Main international conventions relevant to CCS Table A1.1 MARKAL model data structure for a power plant - an exampleTable A1.2 Demand categories in the ETP modelTable A1.3 Coal and gas price projections, 2000-2030 Table A1.4 Key assumptions in the GIS analysis of the potential for renewablesTable A1.5 Damage costs for fossil-fuelled power plants Table A2.1 Region specific cost multipliers Table A2.2 Region and sector specific discount rates in the ETP modelTable A3.1 BASE/GLO50 GDP growth projections (% per year)Table A3.2 BASE/GLO50 per capita GDP Table A3.3 GDP growth projections for the sensitivity analysis with lower growth rates

(% per year)Table A3.4 GDP growth projections for the sensitivity analysis with higher growth rates

(% per year)

007-012 18/11/2004 12:24

OVERVIEW

The IEA World Energy Outlook (WEO) Reference Scenario projects that, based on policies in place,by 2030 CO2 emissions will have increased by 63% from today’s level, which is almost 90%higher than 1990 levels. Even in the WEO 2004’s World Alternative Policy Scenario – which analysesthe impact of additional mitigation policies up to 2030 – global CO2 emissions would increase40% on today’s level, putting them 62% higher than in 1990. Hence, to avoid substantial increasesover the next few decades, stronger actions than those currently being considered by governmentsmust be taken, including the development and deployment of technology options that have thepotential to cut emissions significantly. One such option is to capture the CO2 produced from fueluse at major point sources and prevent it from reaching the atmosphere by storing it.

This study shows that CO2 capture and storage (CCS) is a promising emission reduction optionwith potentially important environmental, economic and energy supply security benefits. But moreresearch and investment into CO2 capture and storage is required. This study highlights the factthat large-scale uptake of capture and storage technologies is probably 10 years off and that,without a major increase in RD&D investment, the technology will not be in place to realise its fullpotential as an emissions mitigation tool from 2030 onwards.

This study compares CCS and other emission mitigation options and assesses its prospects. It describesthe challenges that must be overcome for a CCS strategy to reach market introduction by 2015and achieve its full potential over the next 30-50 years. It identifies the major issues and uncertaintiesthat should be considered when deploying CCS as part of an emission mitigation strategy.

This analysis is in three parts. The first provides a comprehensive overview of the prospects, costsand R&D challenges of CO2 capture, transportation and storage technologies. The secondquantitatively tests the hypothesis that CCS is a viable and competitive strategy for cutting emissionsand that it is worthwhile accelerating RD&D and international efforts to advance CCS to thelevels required. The third highlights the priority actions that would need to be taken for the timelydeployment of CCS as an emissions mitigation tool.

What is CO2 capture & storage?

CO2 capture and storage (CCS) involves three distinct processes, shown in the figure below: first,capturing CO2 from the gas streams emitted during electricity production, industrial processes orfuel processing; second, transporting the captured CO2 by pipeline or in tankers; and third storingCO2 underground in deep saline aquifers, depleted oil and gas reservoirs or unmineable coal seams.All three processes have been in use for decades, albeit not with the purpose of storing CO2.Further development is needed, especially on the capture and storage of CO2. While pipeline transportis an established technology, the siting of CCS projects can reduce the need for an extensivetransportation system. The challenge, cost and environmental impact of such a CO2 pipeline systemshould not be underestimated.

What are the current and planned CCS projects?

An overview of CCS projects is provided in the table below. In most CO2 capture demonstrationprojects, existing technologies are applied. Various small-scale pilot plants based on new capturetechnologies are in operation around the world. Only one power plant demonstration project on amegatonne-scale has so far been announced: the FutureGen project in the US. This is a coal-fired

OVERVIEW 13

013-022 17/11/2004 11:52


CO

2 C

aptu

re,

Tran

sport

an

d S

tora

ge

Con

cept

CA

PTU

RE

for

exam

ple

TR

AN

SPO

RT

STO

RA

GE

for

exam

ple

Pow

er p

lan

tsPip

elin

esSt

ora

ge

in s

alin

e aq

uifer

s

Gas

pro

cess

ing

Ship

sEn

han

ced

oil

reco

very

Illustration: BUG, Statoil

Photo: Sealand, Statoil

Illustration: IEA GHG R&D ProgrammeIllustration: D. Fierstein, Statoil

013-022 17/11/2004 11:52

advanced power plant for cogeneration of electricity and hydrogen. Its construction is planned tostart in 2007. Other demonstration projects are planned in Canada, Europe, and Australia.

There are one hundred ongoing and proposed geologic storage projects. Two of these projectsdeserve special mentioning because of their scale. Storage in deep saline aquifers has beendemonstrated in one commercial-scale project, at the Sleipner site in Norway (sub-sea storage).About 1 Mt of CO2 per year has been stored since 1996. This project is important as it proves thatstorage in aquifers can work in practice. No leakage has so far been detected. Using CO2 to enhanceoil recovery and CO2 storage underground have been demonstrated at the Weyburn project inCanada. About 2 Mt of CO2 per year has been stored since 2001. In both projects the behaviourof the CO2 underground has corresponded to what models had predicted, and important progresswas achieved in the monitoring of CO2 underground. Pilot projects suggest that CO2-enhanced coal-bed methane (ECBM) and enhanced gas recovery (EGR) may be viable but the experience so far isnot sufficient to consider these two as proven options. Encouraged by these promising results, manymore storage demonstration projects have been started or are planned.

Where could CO2 capture technology be applied?In principle, CO2 can be captured from all installations used to combust fossil fuels and biomass, providedthat the scale of the emissions source is large enough. In practice, only three areas are suitable: electricitygeneration (including district heating and industrial combined heat and power generation), industrialprocesses, and fuels processing. Emissions from other sources – such as the transport, agriculture, serviceand residential sectors – are too dispersed to make capture viable. Alternative measures, such as enhancingenergy efficiency, renewables, CHP and increased use of hydrogen produced at centralised facilitiesfitted with CO2 capture technology, may be better options for these sectors.

Since power production is responsible for over 29% of global CO2 emissions, capturing from electricityplants offers the best initial potential for capturing the CO2 generated from fossil-fuel use. To alesser extent, CO2 can also be captured during the production of iron, steel, cement, chemicals andpulp, and from oil refining, natural gas processing and the production of synthetic fuels (such ashydrogen and liquid transportation fuels from natural gas, coal or biomass).

Which CO2 capture technologies are most promising?CO2 can be captured either before or after combustion using a range of existing and emergingtechnologies. In conventional processes, CO2 is captured from the flue gases produced duringcombustion (post-combustion capture). It is also possible to convert the hydrocarbon fuel into CO2

and hydrogen, remove the CO2 from the fuel gas and combust the hydrogen (pre-combustion capture).

OVERVIEW 15

Overview of worldwide CCS projects

No. of projects

CO2 capture demonstration projects 11

CO2 capture R&D projects 35

Geologic storage projects 26

Geologic storage R&D projects 74

Ocean storage R&D projects 9

013-022 17/11/2004 11:52


In pre-combustion, physical absorption of CO2 is the most promising capture option. In post-combustion capture, options include processes based on chemical absorption or oxyfueling(combustion using oxygen separated from air, which generates nearly pure CO2 flue gas). Chemicaland physical absorption are proven technologies. Longer-term, gas separation membranes and othernew technologies may be used for both pre- and post-combustion capture.

In electricity generation, CO2 capture is most effective when used in combination with large-scale,high-efficiency power plants. Indeed, the success of a CCS strategy could depend on the use ofsuch plants. For coal-fired plants, Integrated Gasification Combined Cycle (IGCC) fitted with physicalabsorption technology to capture CO2 at the pre-combustion stage is considered to be promising.Coal-fired Ultra Supercritical Steam Cycles (USCSC) fitted with post-combustion capture technologiesor various types of oxyfueling technology (including chemical looping, where the oxygen is suppliedthrough a chemical reaction), may emerge as alternatives. For natural gas-fired plants, oxyfueling(including chemical looping), pre-combustion gas shifting and physical absorption in combinationwith hydrogen turbines, or post-combustion chemical absorption are promising options. At a laterstage, fuel cells may be integrated into high-efficiency coal- and gas-fired power plants fitted withCCS. Capturing CO2 from plants which co-generate electricity and synthetic fuels could have additionalcost savings compared to stand-alone power production with CO2 capture.

Advances in capture technology are needed to reduce the cost of CO2 capture from power generation.Given the range of ongoing R&D efforts, it is not yet possible to pick a ‘winning’ capture technology.It is likely that several will be used in future. All require further improvements to cut costs andimprove capture efficiency before they can be applied on a commercial scale, a process which islikely to take years. RD&D must be accelerated if CCS is to play a substantial role in the comingdecades and have a significant impact on emissions.

How much CO2 storage capacity is available?

Deep saline aquifers, depleted oil and gas reservoirs and unmineable coal seams offer the bestoption for underground CO2 storage. This includes sub-sea reservoirs. Oceanic storage (i.e., CO2

storage in the water column) is problematic given the unknown environmental impacts. Surfacemineralization is still at a conceptual stage.

In underground reservoirs, CO2 is stored as a bubble under an impermeable caprock at a depth ofmore than 800 meters, in the top part of a water-filled reservoir rock. Deep saline aquifers offerpotentially decades or hundreds of years’ worth of storage capacity with between 1,000-10,000 Gtof capacity available, possibly even more. This is the single most important underground storagepotential. Around 920 Gt of CO2 could be stored in depleted oil and gas fields. The storage capacityof unmineable coal seams, where CO2 is absorbed on the coal surface, is an order of magnitudesmaller. While the absolute value of the potentials are uncertain as of yet, it is clear that they arelarge. CO2 storage may be combined with enhanced oil recovery (EOR), enhanced coalbed methanerecovery (ECBM), and enhanced gas recovery (EGR). Such combinations could create revenues thatmay offset part or even all of the capture and transportation cost.

Many storage sites are far from large emission sources. Coupled with the fact that long-rangeintercontinental transportation of CO2 would incur significant additional cost, this means that theeconomic storage potential is country and region specific and smaller than the total geologic storagepotential. However, in most world regions storage capacities do not pose a constraint for widespreadCCS use for decades to come.

013-022 17/11/2004 11:52

What is the risk and effect of CO2 leakage back into theatmosphere?All three storage options – deep saline aquifers, depleted oil and gas reserves and unmineablecoal seams – need more proof on a large scale. The technology to store CO2 underground shouldbe considered proven technology. The problem is whether the CO2 will leak from undergroundstorage sites back into the atmosphere. The leakage discussion can be split into two parts: thequestion to what extent leakage can reduce the emissions reduction effectiveness of CCS, and publicconcerns that CO2 leakage can be dangerous.

Small leakages of CO2 may occur over a long period of time, which could reduce the effectivenessof CCS as an emission mitigation option. This so-called permanence problem is currently dealt withthrough field tests and through modelling studies. Depleted oil and gas fields have containedhydrocarbons for millions of years. This makes them a relatively safe place to store CO2. The problemfor such reservoirs is mainly if the extraction activity has created leakage pathways, and if abandonedboreholes can be plugged properly so the CO2 cannot escape. The only existing large-scale aquiferstorage demonstration project has shown no leakage since it started eight years ago. Many projectsfor natural gas storage and acid gas storage have worked well. Progress in modelling allowsincreasingly accurate forecasts of the long-term fate of the CO2, which cannot be tested in practice.Several natural phenomena, such as CO2 dissolution in the aquifer water, will reduce the long-termrisk of leakage. The understanding of these phenomena is improving gradually.

CO2 is not toxic, but CO2 can be dangerous in high concentrations as it can cause suffocation dueto lack of oxygen. Accidents where significant amounts of CO2 are released from undergroundreservoirs, with potential risk for local residents, are highly unlikely. The storage under more than800 metres of sediment excludes sudden eruptions of massive amounts of CO2. However, there arecases where natural CO2 emissions from underground have created locally dangerous situations.Proper CO2 monitoring systems and remediation measures can prevent such problems.

While the RD&D results are encouraging, more pilot projects are needed to better understand andvalidate the permanence of underground storage in various geological formations and developcriteria to rank appropriate sites. Too strict criteria for leakage could unnecessarily reduce thepotential for aquifer storage.

What is the cost of capturing, transporting and storing CO2?The future cost of capturing, transporting and storing CO2 depends on which capture technologiesare used, how they are applied, how far costs fall as a result of RD&D (innovation) and marketuptake (learning-by-doing), and fuel prices. Since applying capture requires more energy use andleads to production of more CO2, the cost per tonne of CO2 emission mitigation is higher than theper tonne cost of capturing and storing CO2. The gap between the two narrows as capture energyefficiency increases.

At this stage, the total cost of CCS could range from 50 to 100 USD per tonne of CO2. This coulddrop significantly in future. In most cases, using CCS would cost 25-50 USD per tonne of CO2 by2030, compared to the same process without. Certain early opportunities exist with substantiallylower cost, but their potential is limited.

The cost for CCS can be split into cost of capture, transportation and storage. Current estimatesfor large-scale capture systems (including CO2 pressurization, excluding transportation and storage)are 25-50 USD per tonne of CO2 but are expected to improve as the technology is developed and

OVERVIEW 17

013-022 17/11/2004 11:52


deployed. If future efficiency gains are taken into account, costs could fall to 10-25 USD/t CO2 forcoal-fired plants and to 25-30 USD/t CO2 for gas-fired plants over the next 25 years.

With CO2 transportation, pipeline costs depend strongly on the volumes being transported and,to a lesser extent, on the distances involved. Large-scale pipeline transportation costs range from 1-5 USD/t CO2 per 100 km. If CO2 is shipped over long distances rather than transported in pipelines,the cost falls to around 15-25 USD/t CO2 for a distance of 5,000 km.

The cost of CO2 storage depends on the site, its location and method of injection chosen. In general,at around 1-2 USD per tonne of CO2, storage costs are marginal compared to capture andtransportation costs. Revenues from using CO2 to enhance oil production (EOR) could be substantial(up to 55 USD/t CO2), and enable the cost of CCS to be offset. However, such potential is highlysite specific and would not apply to most CCS projects. Longer-term costs for monitoring andverification of storage sites are of secondary importance.

Using CCS with new coal- and gas-fired power plants would increase electricity production costs by2-3 US cents/kWh. By 2030, CCS cost could fall to 1-2 US cents per kWh (including capture,transportation and storage).

How does the cost-effectiveness of CCS compare to otheremission reduction options? The model analysisCO2 emission reduction options in the energy sector include lower carbon fossil fuels, renewables,nuclear, energy efficiency and CCS. Outside the energy sector there are options such as afforestationand land-use change, and reduction of non-CO2 greenhouse gases. Each option is characterized bya (marginal) cost curve that allows for a certain emission reduction potential at a given CO2 price.Therefore different options co-exist in a cost-effective policy mix. The more ambitious the emissionreduction targets, the more options will be needed, and the more effective and costly the optionsthat will be needed. CCS can reduce emissions by 85 to 95% compared to the same processeswithout CCS but it is a relatively costly emission reduction strategy. Therefore the widespread useof CCS only makes sense in a scenario with significant emission reduction.

The Energy Technology Perspectives (ETP) model is an economic partial equilibrium model. Theworld energy system for the period 2000-2050 is optimized, based on least cost. The model is basedon a detailed representation of the energy system in terms of energy flows and energy technologies.Cost-effective emission reduction options are chosen from a technology database that containsoptions such as CCS, nuclear, renewables and energy efficiency. The model is a suitable tool withwhich to identify the best set of options and to map uncertainties.

CO2 capture and storage (CCS) could potentially allow for the continued use of fossil fuels whileat the same time achieving significant reductions in CO2 emissions. Indeed, the results of IEA analysisshow that CCS could even play a key role in a scenario where global CO2 emissions are roughlystabilized at 2000 level by 2050. This would require significant policy action, however, equivalentto a CO2 penalty level of 50 USD per tonne of CO2. This scenario would halve emissions by 2050compared to a scenario where no additional policy action was taken. CCS technologies contributeabout half of the reductions achieved by 2050.

By 2050, 80% of the captured CO2 would come from electricity production, particularly coal-firedgeneration. At a penalty level of 50$/t CO2, power plants with CO2 capture would represent 22%of total global installed generation capacity by 2050 and produce 39% of all electricity. Withinthe electricity sector, coal-fired IGCCs fitted with CCS that co-generate hydrogen and other

013-022 17/11/2004 11:52

transportation fuels would play an important role. Capture from coal-fired processes would represent65% of the total CO2 captured by 2050, the remainder coming from gas, oil- and biomass-firedprocesses, and from cement kilns.

Up to 2025, CCS would mainly be applied in industrialized countries. By 2050, almost half oftotal capture activity could be rolled out in developing countries, mostly China and India. Technologytransfer from industrialized countries (particularly of efficient power-generation plants) could helpto realize the full potential of CCS in developing countries. If CO2 policies were limited to industrializedcountries, the role of CCS would be significantly reduced. This finding emphasizes the importanceof international co-operation.

What would be the environmental benefits of CCS?The potential benefits of CCS can be further illustrated by considering a scenario without CCS butwith the same emission penalty level (50 USD/t CO2). In this case, emission levels in 2050 wouldincrease by over a quarter compared to the scenario in which CCS was included. In fact, without CCS,the CO2 penalty imposed would need to be doubled before the same reductions could be achieved.

Additional scenarios were analysed that combine various key uncertainties such as the policyambition level, the extent of international co-operation to mitigate emissions and the prospects fortechnological change. These scenarios suggest CCS potentials are between 3 Gt and 7.6 Gt CO2 in2030, and between 5.5 Gt and 19.2 Gt CO2 in 2050. This compares to 38 Gt CO2 emissions by2030 under the WEO Reference Scenario. The fact that all scenarios show a potential on a Gt-scale suggests that CCS technologies constitute a robust option for emissions reduction.

Such results are sensitive to assumptions about future technology development, not only for CCS,but also for other mitigation options such as renewables and nuclear. More optimistic assumptionsfor the future cost reduction of renewables and the potential for expanding nuclear would considerablyreduce the future role of CCS.

One important finding of this analysis is that renewables, nuclear and CCS technologies can co-exist as part of a cost-effective portfolio of options for reducing CO2 emissions from energy production.However, the relative role of each would vary from region to region. It would also depend on policyefforts and cost developments for all technologies, the extent to which promising technology optionsactually work, institutional and legal barriers, and public acceptance (relevant for all three technologyoptions). Investing in CCS RD&D could be a good ‘insurance policy’ for the future. Such a hedgingapproach would reduce the risk of failure.

What would be the fuel market consequences of CCS?CCS would result in a significant increase in the use of coal compared to a scenario where CCS isnot considered, but the same CO2 policies are applied. As coal is considered a more secure fuelthan oil and gas, the fact that coal remains a viable energy option increases supply security. CCSwould have a limited impact on the use of oil and natural gas. CCS would result in a lower use ofrenewables and nuclear and increase clean fossil energy availability. However, this model resultdoes not account for the uncertain growth potential of cost-effective renewables. Coal on the otherhand is an established fuel. As CCS makes coal a more sustainable option, it increases the securityof supply, even in regions where the actual investments in coal are of a limited scale.

For regions with ample coal reserves, such as North America, China and India, CCS could result inlower imports and increased reliance on domestic energy sources. For a number of countries such

OVERVIEW 19

013-022 17/11/2004 11:52


as Australia, coal exports would be higher if global coal consumption were higher. This could haveeconomic advantages that need to be analysed in more detail.

Is CCS relevant for all countries and all regions?The relevance of CCS differs by region. Model analysis suggests that CCS can become an importantoption in North America, Australia and parts of Europe. While the CCS potentials in China andIndia are important as well, the realization of these potentials will depend on the extent of globalefforts to reduce CO2 emissions. If CO2 policies are limited to industrialized countries, the role ofCCS is significantly reduced on a global scale. This finding emphasizes the importance of technologytransfer and international co-operation on both technology and policy.

Given that long-range transportation of CO2 seems an unlikely option given its high cost, for countrieswithout sufficient storage potential close to their emission sources, it may be more cost effectiveto consider alternative emission reduction strategies. While having CCS in a CO2 policy portfoliois certainly attractive, the issue of its application will require a careful case-by-case project evaluation.This evaluation must account for the energy system characteristics on the continental, the countryand the local scale.

What will it take to bring CO2 capture & storage to market?There is a ‘window of opportunity’ for CCS to compete as a technology option, starting fromaround 2020 and peaking in the second half of the 21st century. Beyond that, CO2-free alternativeswould make CCS redundant. In other words, CCS should be considered an essential ‘transitiontechnology’ to a sustainable energy system for the next 50 to 100 years.

The single most important hurdle which CCS must overcome is public acceptance of storing CO2

underground. Unless it can be proven that CO2 can be permanently and safely stored over thelong term, the option will be untenable, whatever its additional benefits.

The potential for 2030 is two to three orders of magnitude greater than the projected Mt-scaledemonstration projects for 2015. This indicates the need for significantly increasing both investmentin RD&D and the scope of projects, if a CCS strategy is to succeed. Taken together, all the plannedCCS projects in the coming decade will barely reach the 10 Mt per year scale. If the full emissionmitigation potential of CCS is to be realized, RD&D activities need to be scaled up and acceleratedsignificantly.

Achieving this will require increasing the number of commercial scale storage pilot projects overthe next 10 years and ensuring that the general public is consulted throughout. RD&D shouldinitially focus on storage projects which enhance fossil-fuel production and those which advanceknowledge on sub-sea underground storage, and aquifer storage in locations with low populationdensity, in order to minimize planning hurdles. Processes which consult, review, comment and addressstakeholder concerns should be built into all pilot projects. Procedures for independently verifyingand monitoring storage and related activities should also be established. Finally, a regulatory andlegal framework for CO2 storage projects must be developed to address issues around liability,licensing, leakage, landowner, royalty and citizens rights.

Governments must address the present shortage of sizeable RD&D projects in order to advancetechnological understanding, increase efficiency and drive down costs. This will require increasingRD&D, investment into CCS demonstration projects, and power-plant efficiency improvements. By2015 at least 10 major power plants fitted with capture technology need to be operating. These

013-022 17/11/2004 11:52

plants would cost between 500 million and 1 billion USD each, half of which would be additionalcost for CCS. The current CCS budget is over 100 million USD per year. The needed RD&D wouldrepresent a fivefold increase. While the amount required is challenging, it is not insurmountablegiven the scale of past energy RD&D budgets. It would represent a 30% increase of the currenttotal RD&D budget for fossil fuels and power & storage technologies. Leveraging the funds inprivate/public partnerships is essential.

Creation of an enabling environment to ensure technology development must be accompanied bythe simultaneous development of legal and regulatory frameworks. In the interests of time, andgiven the diversity of institutional arrangements and policy processes between countries, workingat the national level using existing frameworks may be the best short-term option.

Finally, countries should create a level-playing field for CCS alongside other climate change mitigationtechnologies. This includes ensuring that various climate change mitigation instruments, includingmarket-oriented trading schemes, are adapted to include CCS. The future role of CCS dependscritically on sufficiently ambitious CO2 policies in non-OECD countries. Therefore, outreachprogrammes to developing countries and transition economies and international commitment toreduce CO2 emissions are a prerequisite. The maturation of a global emissions-trading scheme, ameaningful price for CO2 and a predictable return on investment are important factors that couldstimulate the timely deployment of CCS.

OVERVIEW 21

013-022 17/11/2004 11:52

013-022 17/11/2004 11:52

Chapter 1. INTRODUCTION

Governments around the world are increasingly interested in CO2 capture and storage (CCS) as away of mitigating rising greenhouse gas emissions. Various international bodies established underauspices of the International Energy Agency explore the viability of CCS. These include the IEAGreenhouse Gas R&D Programme1 which has been assessing CO2 capture and storage technologyfor over 10 years (Freund and Davison, 2002), and the Working Party on Fossil Fuels, one of theIEA standing committees, which has focused on measures to introduce CO2 capture (McKee, 2002).Such interest is supported by an increasing body of scientific knowledge on CCS collected over thepast 30 years (Marchetti, 1977).2

With energy demand projected to rise by over 60% up to 2030, limiting CO2 emissions fromenergy use is becoming ever more pressing. CCS is a strategy which can ‘buy time’ until CO2-freeenergy solutions prevail. While the projected quantities of CO2 which would need to be capturedand stored to achieve significant global reductions are huge, technologies exist to do so. Theextent to which CCS could be applied, however, remains in doubt, as does the benefit and impactsuch a strategy would have on the world economy, energy markets and electricity prices. Similarly,the rate at which CCS could be adopted and the way this could be done remain unclear, as doesthe attractiveness of CCS compared with other mitigation options, such as renewable energy ornuclear power.

The IEA Governing Board recently emphasized the importance of CO2 capture and storage andrequested the IEA Secretariat to prepare advice on a long-term policy framework to facilitate thecommercial application of capture and storage of CO2 from fossil fuels.

In line with this objective, this study sheds light on the economic potential for CO2 capture andstorage over the next 30-50 years using the Energy Technology Perspectives (ETP) model, aquantitative optimization model developed by the IEA. It assesses the prospects for CCS technologiesbased on the energy resources regions, regional and sectoral shifts in global energy demand, andmodifications in energy technology portfolios. It compares CCS with other emission mitigationoptions and identifies key issues and uncertainties that should be considered in relation to CCSand its use as a CO2 emission mitigation tool.

The study identifies the main uncertainties surrounding CCS, using sensitivity analyses and scenarioanalysis techniques. The results are cross-compared to determine the role for CCS under differentscenarios in key world regions, and analysis is also undertaken to asses the impact of CCS on fuelmarkets using different CO2 abatement incentives.

The ETP model findings are complemented by a detailed description of the current and promisingtechnologies needed for CCS, where these can be applied, and their associated cost and energyrequirements. An assessment is also made of the major policy mechanisms necessary to bring CCS

1. INTRODUCTION 23

1. The IEA Greenhouse Gas R&D Programme is one of 40 IEA Implementing Agreements. The Programme’s main focus is on CO2 captureand storage, making it the world’s leading international research co-operation effort in this field. Some other Implementing Agreements –such as the Clean Coal Centre (CCC), the Energy Technology Systems Analysis Programme (ETSAP), and the Hydrogen ImplementingAgreement – also work in this area, but it is not their main task.

2. Marchetti’s paper focused on power plants with CO2 capture and oceanic storage, using the downward flow from the Mediterraneansea into the Atlantic in the Strait of Gibraltar. Oceanic storage is still regarded as controversial.

023-036 chapter 1-2 18/11/2004 17:26


technologies through R&D and deployment to commercialization, as well as issues around publicacceptance, leakage monitoring, legal and regulatory frameworks, and timing.

The discussion of CO2 capture technologies and the issues surrounding the permanence of CO2 storageaddressed in this book are not exhaustive. The special report on CCS by the Intergovernmental Panelon Climate Change (IPCC), scheduled for release in 2005, will discuss these topics in far greater detail.Nor does this publication aim to provide a technology roadmap. Instead, the reader is referred to thevarious roadmaps published in recent years (e.g. Henderson 2003, CO2CRC 2004, McKee 2004a).

The IEA anticipates that the qualitative and quantitative insights provided by this study will helpgovernments and industries that are considering adopting strategies to mitigate CO2 emissions tobetter understand the status, cost and potential of CCS technologies and the steps required to bringthem to full-scale implementation.

The Purpose of the Study

This book has three purposes. Firstly, it provides a comprehensive overview of the prospects, costsand R&D challenges of CO2 capture, transportation and storage (CCS) technologies. Secondly, ittests the hypothesis that CCS is indeed a viable and competitive option for mitigating CO2 emissionsand that it is worthwhile accelerating R&D and international efforts to advance the technologies.Thirdly, it highlights the actions that would be needed if CCS were to be deployed as part of a CO2

mitigation strategy. The following questions are addressed in this book:

● What are the characteristics of the energy system and its CO2 emissions, and how can the roleof CCS in this system be analysed?

● What is the current status of CCS technologies and what R&D gaps need to be filled?

● How do the environmental and cost benefits of CCS compare with other greenhouse gas emissionmitigation strategies?

● What potential does CCS have as a mitigation strategy, and what are the risks and uncertaintiesof such a strategy?

● What impact would a CCS strategy have on fuel markets and overall energy supply security?

● Is it worth accelerating R&D into CCS technologies and, if so, what are the benefits of doingso and the international efforts required?

The Structure of the Study

The book is divided into eight chapters, four of which present the findings of the Energy TechnologyPerspectives (ETP) model, and two of which provide an overview of the status of CCS technologiesand the major challenges which would need to be overcome if CCS were to be used as part of aCO2 mitigation strategy.

Chapter 2 sets the scene by identifying the factors which have shaped past emission trends andthose likely to be important for the evolution of the future energy system if no further climate orenergy policies are enacted beyond those in place today. This is followed by a discussion of howCO2 emissions could be reduced in the future through the development and deployment of variousnew energy technologies, one of which is CO2 capture and storage (CCS).

023-036 chapter 1-2 18/11/2004 17:26

Chapter 3 provides a qualitative assessment of the technical and economic characteristics of CO2

capture, transportation and storage technologies, data which the ETP model uses to quantify thepotential for CCS. The assessment is based on a comprehensive review of governmental publications,industry studies, peer-reviewed scientific literature and ‘grey’ literature such as workshop presentations.

Chapters 4, 5, 6 and 7 set out the four groups of results from the ETP model analysis.

Chapter 4 begins with an overview of the ETP model, setting out the method used to quantitativelyassess the way in which CCS could reduce global CO2 emissions. It then gives the results from theETP BASE Scenario, followed by a detailed analysis of one scenario in which a penalty of 50 USD/tCO2 is imposed globally, a reference case known as the GLO50 Scenario. Finally, it considers thebenefits of deploying CCS, based on a set of model runs with and without CCS.

Chapter 5 discusses the results of sensitivity analysis undertaken on the ETP model’s GLO50 Scenarioin order to map the various uncertainties associated with individual parameters for CCS technologies,or parameters that affect the use of CCS technologies. Understanding this is a key part of determiningif and how a CCS strategy should be applied, the impact that factors such as economic growth,technology development, environmental policy and regional deployment could have on CCS use,and the interrelation between the parameters.

Chapter 6 presents and compares the results of four ETP scenarios to assess the interactions ofspecific parameters identified during the sensitivity analysis outlined in Chapter 5 and hence therobustness of the results. In the first instance, the scenario results for CCS are considered on a globallevel. This is followed by a discussion of the regional scenario results. The regional results are thencompared against actual and planned RD&D activities. This analysis provides insights for the CCSpolicy challenges discussed in Chapter 8.

Chapter 7 discusses the consequences of deploying CCS on fuel markets based on ETP model analysis.Apart from environmental concerns, supply security and economic consequences play an importantrole in the design of energy policies. While the analysis in the previous chapters showed that a CCSstrategy can result in a significant reduction of CO2 emissions and also lower the cost of environmentalpolicies, it is less clear what impacts CCS would have on supply security and fuel prices.

Chapter 8 outlines the additional uncertainties associated with deploying CCS which are outside thescope of the ETP model but which could critically impact the timing and effectiveness of a CCS strategy.These factors, which are strongly interrelated, include bridging the RD&D gap to realize the potentialfor CCS, the need for public awareness and acceptance, the importance of putting in place appropriatelegal and regulatory frameworks, particularly for CO2 storage, and the need for a policy frameworkwhich encourages public-private sector co-operation and provides appropriate investment incentives.

1. INTRODUCTION 25

023-036 chapter 1-2 18/11/2004 17:26

023-036 chapter 1-2 18/11/2004 17:26

Chapter 2. THE WORLD ENERGY SYSTEM, CO2EMISSIONS AND MITIGATION OPTIONS

This chapter briefly reviews historical and future baseline developments of global CO2 emissions.It sets out the factors which have shaped past emission trends and those likely to be important forthe evolution of the future energy system if no further climate or energy policies are enacted beyondthose in place today. This is followed by a discussion of how CO2 emissions could be reduced inthe future through the development and deployment of various new energy technologies, one ofwhich is CO2 capture and storage (CCS).

Global CO2 Emissions: Past Trends and Future Outlook

Global CO2 emissions increased by over 70% between 1971 and 2002. Since 1990 CO2 emissionshave risen by some 16% (Figure 2.1). The Reference Scenario of the IEA World Energy Outlook (WEO)projects that global emissions will be up 63% on today’s level by 2030, around 90% higher than 1990levels (IEA, 2004a). This corresponds to an average growth rate of 1.7% per year, roughly the same rateas over the last three decades (IEA, 2004b).

Historically, CO2 emissions have come overwhelmingly from industrialized countries. However, two-thirds of the increase up to 2030 is expected to come from developing countries. By 2030, developingnations are set to account for almost 49% of global CO2 emissions (up from 35% today), withOECD countries accounting for 42% and transition economies for 9% (IEA, 2004a).

2. THE WORLD ENERGY SYSTEM, CO2 EMISSIONS AND MITIGATION OPTIONS 27

Figure 2.1

Energy-related CO2 emissions, globally and by region (1973-2030)

Gt

CO

2 p

er y

ear

40

35

25

30

20

15

10

5

0

2030

2020

2010

2000

1990

1980

1970

Transition Economies

OECD

Developing Countries

World

Note: Excludes CO2 emissions from international marine bunkers.

Source: IEA, 2004a.

023-036 chapter 1-2 18/11/2004 17:26


Over the past three decades, the burning of coal accounted for 40% of the increase in global CO2

emissions, with oil responsible for 31% and gas 29%. The WEO Reference Scenario projects thatof the increase in emissions between 2002 and 2030, oil will account for 37%, coal 33% and gas30% (Figure 2.2). As a result, the share of coal in total emissions will drop from 39% in 2002 to36% in 2030. The share of gas will rise from 21% to 24%, while the share of oil in total emissionsremains roughly unchanged over the outlook period.

The Intergovernmental Panel on Climate Change (IPCC) published a Special Report on EmissionScenarios (SRES) in 2000 setting out a range of global emission paths up to 2100 (Nakiçenoviçet al., 2000). Figure 2.3 shows CO2 emissions from energy use under three of these scenarios, allof which form part of the so-called SRES ‘A1’ storyline. These scenarios describe a future with veryrapid economic growth in which the global population peaks in mid-century and declines thereafter,and in which new, more efficient technologies are rapidly introduced.

The A1 scenario group is divided into three sub-sets which describe alternative directions fortechnological change in the energy system: fossil-intensive (A1F1), non-fossil energy sources (A1T);and a balance across all energy sources (A1B).

While the level of CO2 emissions in the three scenarios are fairly evenly matched up to 2020, theyall show higher emission levels than the WEO 2004 Reference Scenario. Emissions increase through2050 in all scenarios with 2050 levels ranging from 45 Gt to 85 Gt CO2. Major differences occurafter 2050. The cumulative emissions in these three scenarios are such that the CO2 concentrationswould range from 600 to 950 ppm by 2100, compared to the current concentration of roughly375 ppm. This would result in an average global temperature increase of 3, 3.5 or 5 degreesCelsius for A1T, A1B and A1F1, respectively (with a margin of error of ±25%).

Figure 2.2

Energy-related CO2 emissions by fuel (1973-2030)

Gt

CO

2 p

er y

ear

16

14

10

12

8

6

4

2

0

2030

2020

2010

2000

1990

1980

1970

Gas

Oil

Coal

Note: Excludes CO2 emissions from international marine bunkers.

Source: IEA, 2004a.

Even the A1T scenario, in which renewable energy technologies significantly reduce emissionsafter 2050, would result in significant global warming. This highlights the importance of reducingCO2 emissions early to avoid atmospheric concentrations reaching levels that would have a seriousimpact on the climate beyond mid-century.

Factors Affecting and Strategies to Reduce CO2Emissions

Both the WEO Reference Scenario and the IPCC scenarios paint a challenging picture for energypolicy makers: unless much stronger action is taken, CO2 emissions from the global energy systemwill continue growing, with potentially serious implications for global warming. This implies thatdeep cuts in CO2 emissions will only come about by transforming the way in which energy is suppliedand used. While governments may emphasize different elements in their emission reduction strategies,this transformation must involve the more efficient production of fuels and electricity, the use ofcleaner fuels and improved efficiency in converting energy into services for end-use consumers.

Understanding the factors which have shaped CO2 emissions in the past, and those likely to do soin the future, is a first step in defining an emission reduction strategy. Growth in CO2 emissions isclearly linked to the use of fossil fuels to meet the ever-increasing demand for energy services. Inturn, demand for energy services is driven by economic growth, although historical data show thatthere has been a decoupling of the two: OECD countries and many non-OECD countries have


Figure 2.3

CO2 emissions in the IPCC SRES A1 scenarios, compared to the WEOReference Scenario

Gt

CO

2 p

er y

ear

120

100

60

80

40

20

0

2100

2080

2090

2070

2060

2050

2040

2030

2020

2010

2000

1990

IPCC A1T

WEO RS

IPCC A1B

IPCC A1F1

Source: IEA, 2004a; Nakićenović et al., 2000.

023-036 chapter 1-2 18/11/2004 17:26


generally been successful in reducing the need for energy to fuel their economies. Today it takesonly 70% of primary energy (in TPES terms) to produce one unit of the world’s GDP than it didthree decades ago. This decoupling of TPES from GDP can primarily be explained by three factors:

● Reduced demand for energy services relative to GDP;

● Improved energy efficiency in end-use sectors;

● Improved supply-side efficiency, particularly in electricity generation.

When considering the factors which link growth in GDP and energy-related CO2 emissions, the listcan be expanded to include the following:1

● The carbon content of the fuel used in different end-use sectors;

● The carbon content of the fuel used in electricity and district heat generation.

Average growth rates of world GDP and CO2 emissions from 1973 to 2002 and throughout theprojection period of the WEO are illustrated in Figure 2.4. Over the last three decades, GDP grewby 3.1% per annum, while CO2 emissions increased at an average annual rate of 1.5 %, indicatingan average annual decoupling rate of 1.6%. World economic growth throughout the Outlook periodis assumed to be close to that seen from 1973-2002, but with a slightly stronger increase in CO2

emissions of 1.7% per annum.

Figure 2.4

Growth in GDP and CO2 emissions: decomposition of factors affecting the link

2002-20301973-20024

3

2

1

0

– 1

– 2

– 3Ave

rag

e %

Ch

ang

e p

er y

ear

GDP

CO2/kWh Power Generation

End-use fuel mix

TFC/GDP

CO2

Source: IEA, 2004a.

1. In theory, the carbon content of fuels used to produce other secondary energy carriers (hydrogen and synthetic fuels) should also betaken into account. However, this factor has played a minor role in the past due to very modest production levels of these fuels to date.

023-036 chapter 1-2 18/11/2004 17:26

Figure 2.4 also illustrates the way in which CO2 emissions have been decoupled from economicgrowth. The third bar in the figure illustrates the impact that changes in total final energy consumption(TFC) per GDP (factors 1 and 2 above), end-use fuel mix (i.e., final energy mix, factor 4) and CO2

intensity in power generation (factors 3 and 5) have each had on total CO2 emissions developments.By far the most important component is the reduction in TFC per GDP. This decline is primarily dueto improved end-use energy efficiency, although structural changes reducing the need for energyservices relative to total GDP (e.g., increased GDP from the service sector relative to that from steelproduction) also affected the development. A similar trend is expected up to 2030.

The world end-use fuel mix (including upstream emissions in electricity production) has becomemarginally more CO2 intensive in recent decades, mostly due to the higher share taken up byelectricity. Worldwide, electricity has a higher CO2 intensity than fossil fuels due to generation lossesand the high share of coal in the generation mix in most world regions (Figure 2.5). With a steadyincrease in the share of electricity in global final energy consumption expected up to 2030, theend-use fuel mix will continue to be a driving force for growth in global CO2 emissions. On theother hand, the CO2 intensity reduction of power generation itself has contributed to a loweringof emissions relative to GDP. This trend is expected to continue over the next 2-3 decades, althoughto a lesser extent.

In Figure 2.6, the fourth bar of each time period illustrates the impact of different factors affectingemissions from power generation; changes in the share of renewables; changes in the share ofnuclear; changes in the mix of fossil fuels used for power generation; and changes in the efficiencyof fossil-fuel based generation. The average annual percent change in these components adds upto the average annual percent change in CO2 emissions per unit of electricity produced. The figureshows that the expansion of nuclear energy is the main reason for an historic decline in CO2 intensity


Figure 2.5

CO2 intensity in global electricity generation and fossil fuels

9

8

7

6

5

4

3

2

1

0Mt

CO

2/M

toe

1973

2002

2030 Coa

lOil

Gas

Average carbon content for each fuel in 2002

Electricity

Source: IEA, 2004a.

023-036 chapter 1-2 18/11/2004 17:26


in power generation. On the other hand, the share of renewable energy in power generation actuallyfell between 1973-2002, while more efficient fossil-fuelled power plants helped cut emissions overthis period. The impact of changes in the fossil-fuel mix was modest.

In the WEO Reference Scenario, the reduction in carbon intensity in power generation over theOutlook period is less significant than that seen during the previous three decades. The main reasonfor this is that the share of nuclear energy in global electricity generation declines over the Outlookperiod. Although a substantial improvement in fossil-fuel-based generation efficiency and a lowercarbon-fuel mix help to reduce emissions per unit of electricity produced, the total result is less ofa decline in CO2 intensity than that seen between 1973 and 2002.

The increasing and then declining role of nuclear energy in the global electricity mix is also illustratedin Figure 2.7. The historical decline in the share of hydropower is expected to continue throughoutthe Outlook period. The increased share of other renewables, mostly wind, more than compensatesfor this decline so that the total share for renewables is slightly higher in 2030 than in 2002 (alsoindicated in Figure 2.6).

Coal has generally maintained its role in global electricity generation and is expected to do so over theOutlook period. Thus, the impact of changes in the fossil fuel generation mix shown in Figure 2.6 canbe explained by natural gas taking a market share from oil. Gas has a somewhat lower carbon contentthan oil, which in turn has a lower carbon content per unit of energy than coal (see Figure 2.5).

Based on the decomposition analysis discussed above, Table 2.1 indicates the relative importancevarious factors have had on changes in past CO2 emission level and on expected future emissions,based on the WEO 2004 Reference Scenario.

Figure 2.6

Growth in electricity demand and CO2 emissions from power generation:decomposition of factors affecting the link

2002-20301973-20024.0

3.0

3.5

2.0

2.5

1.0

1.5

0.0

0.5

– 1.0

– 0.5

– 1.5Ave

rag

e %

Ch

ang

e p

er y

ear

Electricity demand

CO2/kWh from PG

PG renewable share

PG nuclear share

PG efficiency

PG fossilfuel mix

CO2 from PG

Note: PG = power generation.

Source: IEA, 2004a.

023-036 chapter 1-2 18/11/2004 17:26

The different factors listed in Table 2.1 are important to understanding how emissions have changedin the past, how they may change in the future and thence how they may be reduced through policyintervention. Examples of energy and environmental policies targeted at these components include:2

● In transport, policies which encourage a shift away from private vehicle use to less energy-intensive public transport options;

● In end-use sectors, policies to improve energy efficiency such as standards and labelling forelectric appliances, and voluntary agreements with industry;


Figure 2.7

Share of fuels in global electricity generation

100

80

60

30

90

70

50

10

40

20

0Perc

ent

(%)

1973

2002

2030

Nuclear

Hydro

Other renewables

Gas

Oil

CoalCoal

Source: IEA, 2004a.

Note: +, ++ and +++ denotes small, medium and large positive impacts on emission reductions, and - denotes small negative impacts;0 denotes a neutral effect.

Table 2.1

Impact on global CO2 emission reductions from different factors

Factor 1973-2002 2002-2030

Energy services per GDP + +

End-use efficiency +++ +++

Generation efficiency + ++

End-use fuel mix – –

Generation fuel mix:

- Fossil fuels 0 +

- Nuclear ++ –

- Renewables – +

2. In addition to energy policy initiatives, the risk of global warming could also be mitigated by reducing emissions of non-CO2 greenhousegases, by minimizing tropical deforestation, and by sequestering carbon in trees and soil.

023-036 chapter 1-2 18/11/2004 17:26


● In fossil-fuelled power generation, incentives to increase efficiency by developing advancedtechnologies, such as Coal Integrated Gasification Combined Cycle plants (IGCC) and ultrasupercritical steam cycles (USCSC);

● Policies which encourage the use of biomass for heating and industrial use, and the use ofnatural gas as a ‘cleaner’ substitute for coal;

● In power generation, policies which influence fuel mix, for example to enable the developmentand deployment of renewable technologies, to allow for or negate the use of nuclear powerand in some cases, to promote the use of natural gas instead of coal.

The World Energy Outlook 2004 includes a World Alternative Policy Scenario that analyses theimpact policies that are being considered by OECD countries and other countries can have ondevelopments through 2030. The policies analysed include polices in all the categories mentionedabove. As a result of these policies, global primary energy use would be 10% lower in 2030 thanin the Reference Scenario and the energy-related CO2 emissions would be 16% lower. Almost60% of the cumulative reduction in emissions will occur in non-OECD countries, reflecting primarilyhigher potential for efficiency improvements in transition and developing economies.

The difference between the rates of CO2 emission growth in the WEO Reference and AlternativePolicy scenarios is summarized in Figure 2.8. Almost 60% of the worldwide difference is the resultof end-use efficiency measures encouraging the uptake of efficient vehicles, stricter standards forbuildings, and appliances, and more efficient industrial processes. In transition economies anddeveloping countries, the role more efficient played by energy efficiency measures is particularlylarge and reflects the potential for efficiency improvements. The other big contributor to lower

Figure 2.8

Reduction in energy-related CO2 emissions in the WEO Alternative PolicyScenario, by contributory factor

100

80

60

30

90

70

50

10

40

20

0Perc

ent

(%)

Wor

ldOEC

D

Develo

ping

coun

tries

Tran

sition

econ

mies

Changes in the fossil-fuel mix in power generation

Increased nuclear in power generation

Increased renewables in power generation

Fuel switching in end uses

End-use efficiency gains

Note: PG = power generation.

Source: IEA, 2004a.

023-036 chapter 1-2 18/11/2004 17:26

emissions is the increased share of renewable energy in power generation which accounts for a20% reduction in global CO2 emissions. An increased role for nuclear accounts for an additional10%. Fuel switching in end-uses and switching from coal to natural gas in power generation canexplain the rest.

Both historical developments and the WEO 2004 projections indicate that end-use energy efficiencyis the most import factor affecting the decoupling of CO2 emissions from economic growth. However,a recent IEA study of energy use and CO2 emissions in IEA countries showed that the rate ofenergy efficiency improvements has slowed significantly since the late 1980s (IEA, 2004b). This isa general trend across all sectors and in almost all countries. While the economy-wide energy savingsrate for the group of OECD countries included in the study averaged 2% per year between 1973-1990, it had fallen to just 0.7% per year by the end of the 1990s. This has implications for CO2

emissions: a slowdown in the rate of energy saving is the primary reason for the weaker decouplingof CO2 emissions from GDP growth observed in most OECD countries since 1990.

The IEA study’s findings indicate that the oil price shocks in the 1970s and resulting energypolicies did considerably more to limit growth in energy demand and CO2 emissions in OECDcountries than the energy efficiency and climate change policies implemented in the 1990s. Asdemonstrated in the WEO Alternative Policy Scenario, there is still considerable potential for improvingenergy efficiency, although recent trends indicate that OECD governments must make a strongereffort than in the past to exploit what potential remains.

Even with the policies analysed in the WEO Alternative Policy Scenarios, global emissions of CO2

would increase by 37% on today’s level, compared to 63% in the Reference Scenario, puttingthem almost 60% higher than 1990 levels. This implies that, in order to avoid substantial increasesin emissions over the next few decades, options which cut emissions from fuel and electricitysupply must be pursued in addition to improved energy efficiency in end-use sectors.

One such option is CO2 capture and storage (CCS) which could reduce emissions while still allowingfor continued fossil fuel use. Applying CCS in the electricity sector would reduce the carbon intensityof generation; it can be used in the production of synthetic fuels to provide low (or zero) carbonfuels for end-use sectors; it can also be applied to manufacturing processes to reduce the carbonintensity of this sector.

CCS competes and interacts with the other strategies discussed above. Low-cost options to reducenon-CO2 greenhouse gases reduce the need for CO2 emission reductions. If electricity consumptiondeclines because of electricity savings, the potential for reducing emissions from electricity productionalso declines. Other emission abatement costs are influenced by this interaction. For example, thecost of biomass feedstock for transportation fuels depends on biomass demand for power generation.

The characteristics and potential of CO2 emission mitigation options also differ by region, differenceswhich must be taken into account when emission mitigation potentials are assessed. In industrializedcountries, advanced technologies and capital intensive processes may be favoured. In developingcountries, labour-intensive processes may stand a better chance. The potential for energy efficiencyimprovements is generally higher in developing countries than in developed countries.

All of this is further complicated by technological change. RD&D result in new technologies becomingavailable, while investments can result in further cost reductions due to learning-by-doing. Thetechnologies that will be available 30 years from now could be radically different from those inplace today. Proper consideration of technological change is essential for assessing the potentialof and comparing long-term emission mitigation strategies.


023-036 chapter 1-2 18/11/2004 17:26


As has been shown above, changes in CO2 emissions are a function of many factors, with technologicalchange playing a key role. While developments in the short and medium term can be projectedwith a reasonable degree of certainty, uncertainty increases for longer-term projections of half acentury or more. Technological change is one of the uncertain drivers in this timeframe. A rigidanalytical modelling framework can help assess emission reduction strategies and future trends inenergy use in a structured, reproducible and logical way. For this reason, part of this study is basedon modelling analysis.

The tool used for the analysis is the IEA’s newly-developed Energy Technology Perspectives (ETP)model. This global 15-region model allows for an analysis of fuel and technology choices throughoutthe energy system, from the extraction of energy sources, via fuel conversion and electricity generationto technologies in all end-use sectors. The model’s detailed representation of technology optionsincludes several hundred technologies in each of the regions covered by the model. The ETP modelbelongs to the MARKAL family of bottom-up modelling tools. ETP-MARKAL is a linear programmingmodel that minimizes total energy system costs over a 50-year period, given certain levels ofenergy service demands and constraints, such as the availability of natural resources.

This modelling analysis helps assess the cost implications of different CO2 mitigation strategies.While technological change holds the key to meeting the world’s future energy needs simultaneouslycapping emissions, the change will not come about voluntarily or without cost. Governments mustbe willing to encourage the development and deployment of clean and efficient technologies evenif the up-front costs are not negligible.

The model analysis in this study uses CO2 penalty levels as a representation of policy efforts thatgovernments may take to reduce CO2 emissions from energy production and use. This does notimply that CO2 taxes should be the preferred policy measure; penalties are merely a way ofrepresenting the stimuli needed to bring the technologies that can cut emissions to market. Thisapproach ensures that the technology mix in each scenario is selected on the basis of equivalentemission mitigation costs.

The uptake of technology options in the different ETP scenarios presented in this book depends onthe cost and performance assumptions made for each technology. Chapter 3 discusses in detailthe assumptions made for CSS technologies included in this study. Assumptions for other keytechnologies are presented in Annex 1.

023-036 chapter 1-2 18/11/2004 17:26

Chapter 3. CCS CHARACTERISTICS: TECHNOLOGIES,POTENTIAL, PERMANENCE AND COST

H I G H L I G H T S

CCS can build on existing technologies. It involves the separation of CO2 produced duringfossil fuel use, its transport, and its storage, e.g., in geological media. All three activities havebeen implemented on a commercial scale in certain applications. Compared to many otherCO2 abatement options in the power sector, CCS requires less restructuring of energy supplysystems and even a few projects could have a noticeable impact on country-level emissions.

CO2 Capture Sources

The following three areas offer the best potential for large-scale, centralized capture of CO2:

■ Electricity generation. With fossil-fuelled power production responsible for 29% of totalcurrent CO2 emissions, capturing CO2 from coal, natural gas, oil and biomass-fired powerplants is the most promising area in which to apply CCS technology.

■ Industrial processes. CO2 can be captured from the production processes of iron, cement,chemicals and pulp, activities which generate a combined 23% of world CO2 emissions.In some cases, the cost of applying CCS in industry is lower than for power generation;in other cases it is similar.

■ Fuels production. CO2 can be captured from oil refineries, natural gas processing installationsand synfuel production. This includes hydrocarbon synfuels and hydrogen. The co-productionof electricity and synfuels with CO2 capture would result in cost savings per tonne of CO2

captured, compared to stand-alone electricity production.

CO2 Capture Technology

■ Currently available technologies can be used to either de-carbonise fossil fuels to producehydrogen (pre-combustion capture), or to capture CO2 from flue gases (post-combustioncapture). For established technologies, gaps include improved specialised chemical andphysical solvents to decrease energy requirement of capture process. For novel processes,investigations are focused on better and cheaper membranes to increase CO2 concentration,more efficient air separation technologies (some options involve combustion in pureoxygen), cheaper and more efficient fuel cells (to convert chemical energy stored in hydrogenor methane into electricity), hydrogen turbines, chemical looping and others.

■ In order to be applied by the power sector, capture technologies need, in addition tobasic R&D requirements, to be demonstrated on a much bigger scale than has so farbeen required by the chemical industry. They should be optimized together with highlyefficient power plant technologies.

3. CCS CHARACTERISTICS: TECHNOLOGIES, POTENTIAL, PERMANENCE AND COST 37

037-100 chapter 3 18/11/2004 07:20

■ CO2 capture is energy intensive and results in increased coal and gas use for electricityproduction. The increase ranges from 39% for current designs to 6% for advanced futuredesigns. Energy efficient power production is a prerequisite for CCS use in the power sector.

■ Retrofitting high efficiency gas-fired power plants may be a feasible option in the future,if gas prices are sufficiently low. Pulverised coal-fired plants could also be retrofitted, withoxyfuelling seeming the best option.

■ Reducing the cost of CO2 capture is a major factor influencing the long-term viability ofCCS. Several technologies already under development could reduce the cost of CO2 capturebut this requires concerted RD&D. The cost reduction potential through innovation seemssignificant, so efforts should initially aim for RD&D, rather than investment programmes.

■ Emerging technologies such as membrane separation, oxyfuelling in combination withnew oxygen production technologies, chemical looping and fuel cells hold promise ofcutting energy use and cost in half. Currently, it is not possible to identify the ‘winningtechnology’.

■ Combining biomass-fuelled IGCCs with CCS could become attractive, even on a muchsmaller plant scale than coal-fuelled IGCCs due to the doubled CO2 benefits comparedto coal with CCS. Such a combination would have negative net emissions, since biomasscarbon is based on CO2 that plants have captured from the atmosphere. Biomass couldbe used in dedicated plants or, more likely, it could be co-combusted in fossil-fuelled plants.

CO2 Transportation Technology

■ While pipeline transport is an established technology, the proper siting of CCS projectscan reduce the need for an extensive transportation system. Given potential pipeline sitingconstraints and transportation distances of hundreds of kilometers, a CO2 transportation‘backbone’ may be needed to which multiple power plants and a number of storage sitescan be connected. Such a system would allow transportation over longer distances atacceptable cost. Transporting CO2 by ship is also considered as an option.

CO2 Storage Technology

■ Underground CO2 storage in deep saline aquifers, in depleted oil and gas reservoirs andin un-mineable coal seams seems the only realistic option in the short and medium term,due to reasons such as environmental risk and acceptance problems for oceanic storage,as well as cost and immature technology status for above-ground carbonate storage.

■ Widespread use of CCS implies storage in deep saline aquifers. Deep saline aquiferspotentially offer decades or hundreds of years’ worth of storage with between 1,000 and10,000 Gt of storage capacity available, possibly even more. Aquifers are more evenlydistributed than oil and gas reservoirs.

■ Depleted oil and gas reservoirs offer considerable potential as CO2 storage facilities whichcould hold decades of global CO2 emissions with a reasonable degree of certainty.


037-100 chapter 3 18/11/2004 07:20

■ Injecting CO2 to enhance the recovery of fossil fuels could become a key early CO2 storageopportunity, particularly since it can generate revenues that offset all or part of CO2 captureand transportation costs.

■ Several projects in a range of countries have proven the viability of storing CO2 underground.Such projects include injecting CO2 to enhance oil recovery and acid gas injection. However,this potential is not evenly distributed around the world.

Permanence of Storage and Monitoring

■ There are two types of risk associated with leakages of CO2: local, i.e., site specific, affectinghealth, safety and environment, and global, resulting from a return of stored CO2 to theatmosphere. Considering only the latter, i.e., taking the storage effectiveness point of view,leakages of up to 0.1%/year seem acceptable. Maximum allowable leakage rates will set anupper bound on CO2 losses in permit and accounting procedures, but this does not mean thatthe research community expects such leakages which, in reality, should be many times smaller.

■ More pilot projects are needed to assess the permanence of aquifer storage, develop criteriaagainst which the most suitable sites can be selected, and establish adequate monitoringprocedures.

The Cost of CCS

■ The cost of CCS ranges from a 40 USD benefit to a 100 USD/t CO2 cost, if all capture andstorage options are considered. At this stage, for a vast majority of options, the total cost ofCCS could be within 50 to 100 USD per tonne of CO2 emission reduction. By 2030, thesecosts should go down to 25-50 USD per tonne of CO2 compared to the same process withoutCCS. Certain early opportunities exist with low-capture cost, but their potential is limited.

■ The bulk of costs is on the capture side. If future efficiency gains are taken into account,the cost of capture could decline from the current level of almost 50 USD/t CO2 to around10-25 USD/t CO2 for coal-fired plants and to around 25-30 USD/t CO2 for gas-fired plants.

■ CO2 transportation costs (per t of CO2) depend strongly on the volumes being transportedand, to a lesser extent, on the distances involved; They range from 1 to 10 USD/t of CO2,provided the pipeline transports more than 1 Mt of CO2 per year and the distance is lessthan 500 km.

■ CO2 injection costs range from 2 USD/t CO2 (for Mt size aquifer storage) to 50 USD forcertain ECBM projects.

■ Part of the cost of CCS could be offset by revenues from enhanced fossil-fuel production. Thesebenefits could reach 55 USD/t CO2. Revenues from enhanced oil recovery (EOR) in particularcould be substantial, but this is highly site-specific and will not be the case for most CCS projects.

■ Emerging technologies could result in a CCS electricity cost increase of only 1-2 US centsper kWh (including capture, transportation and storage). This cost range applies to bothcoal and gas-fired plants.


037-100 chapter 3 18/11/2004 07:20

This chapter provides an assessment of the technical and economic characteristics of CO2 capture,transportation and storage technologies, data used in later chapters of this book to quantitativelyassess CCS potentials using the IEA’s Energy Technology Perspectives (ETP) model. The analysis isessential reading for policy makers wishing to understand the range of CCS technologies on offer,their relative costs, merits and technology gaps and the areas in which further RD&D is required.

The viability of a CCS strategy depends on several key factors – the cost of CO2 capture, transportationand storage, the potential for CO2 storage and the permanence of storage – all of which arediscussed in this chapter. The discussion will show that although capture costs dominate theoverall CCS cost, there is potential for improvement. Developments are being made on a numberof new CO2 capture technologies that can reduce energy consumption and capture cost. R&D is akey to these cost reductions since the potential for learning-by-doing seems limited. However, atthis stage it is not yet possible to select winning technologies.

The chapter begins with a detailed overview of CO2 emission sources showing which sectors offerthe best potential for applying CO2 capture and storage (CCS). An assessment is then made ofCO2 capture technologies, both existing and speculative, in power generation, manufacturing andfuel processing. It includes a discussion on why CO2 capture only makes sense for high efficiencypower plants, the role of decentralized generation and cogeneration and technology learning effects.

CO2 transportation and the benefits of producing chemicals and fuels from CO2 are then examined.This is followed by assessment of CO2 storage options, from aquifers to depleted oil and gas-fields,including issues surrounding storage permanence and monitoring. Special attention is given tothe use of CO2 to enhance fossil fuel production, which could generate revenues to offset all orpart of CO2 capture costs. Finally, the overall cost-effectiveness of CCS is assessed and the impactits deployment could have on electricity prices.

General Characteristics of CO2 Capture and Storage

CO2 capture and storage involves the separation of CO2 produced during fossil fuel use, its transport,and its storage, e.g., in geological media. All three activities have been implemented on a commercialscale in certain applications. For example, CO2 capture is widely used in the chemical industry. Likewise,pipeline transport of CO2 is an established technology, and CO2 storage is used for enhanced oil recovery.These technologies must be further developed and demonstrated if they are to become a feasibleoption on the scale required. Before the characteristics of current and future technologies are discussed,a number of important features of a CCS strategy will be considered. These strategy features are thestarting point for a discussion as to whether CCS is feasible.

CO2 is the most important anthropogenic greenhouse gas. Over the past 200 years its concentrationin the atmosphere has increased from 0.0275% to 0.0370%. This is largely a result of the combustionof fossil fuels. One way to reduce CO2 emissions is to capture CO2 before it is emitted into the atmosphereand store it elsewhere. In order to mitigate the risk of global climate change, huge quantities of CO2

need to be captured and sequestered. In 2000, 23.4 gigatonnes1 of CO2 were emitted worldwide.2 Ata density of 500 kg/m3, total annual global emissions could be contained in a cube measuring 3.5 kmin length, width and height. Although this is a large volume, it is not impossible to store quantities ofthis order of magnitude.

1. A gigatonne is a billion tonnes, equal to 109 tonnes

2. Energy related emissions including international bunkering. Excludes cement production and tropical deforestation.


037-100 chapter 3 18/11/2004 07:20


If gaseous CO2 is pressurized, it either becomes liquid or reaches a dense state called ‘supercritical’, astate between gas and liquid. The supercritical state occurs at temperatures greater than 31.1°C andpressures greater than 7.38 MPa3. The density of liquid and supercritical CO2 varies with the pressureand temperature, from 200 to more than 1000 kg/m3 (Bachu, 2000). These physical properties influencethe options for capturing, transporting and storing CO2.

A CCS strategy can build on established technologies. CO2 capture has been widely applied in industrialprocesses for decades. CO2 needs to be removed from gas streams in a number of processes such as theproduction of hydrogen, ammonia and Direct Reduced Iron (DRI). The total quantity captured is in therange of 100 to 200 Mt CO2 per year. CO2 capture is also applied in the processing of natural gas.Some of this captured CO2 is transported and used in the production of urea fertilizer and carbonatedbeverages, but most of it is vented. Around 40 Mt of CO2 per year is also extracted from naturalunderground reservoirs and transported over hundreds of kilometers, to be used for enhanced oil recovery(EOR).

Existing, proven CO2 processing technologies have not been developed for the purpose of CO2 emissionmitigation. While experience with these processes shows that the principle works on a large scale, aCCS strategy will require much new technology backed by a substantial reduction in the overall cost ofCCS. The technology status and outlook for CCS will be discussed in more detail later in this chapter.

Underground storage of CO2 in deep saline aquifers has been demonstrated in one commercial scaleproject, the Norwegian Sleipner facility. Other projects have only just started. While further effort isneeded to demonstrate safety and to better understand the permanence of underground storage invarious geological formations, deep saline aquifers represent a potentially huge and widely availablemedium for CO2 storage.

CCS fits into the existing technology trajectory of fossil-fuel based energy supply and can be developedby existing energy technology suppliers. No major adjustments are needed in the energy infrastructure.This would avoid risky transition and unpopular economic restructuring efforts. Key players such as thefossil fuel industry and power producers have expressed their interest in a CCS strategy. Such supportmakes the strategy more attractive from a policy maker’s perspective.

CCS allows the use of coal resources while reducing CO2 emissions dramatically, compared to fossil-fuelled power plants without CO2 capture. The use of coal instead of oil or gas may have importantsupply security benefits. These CCS advantages are lacking for many of the competing strategies.

In principle, CO2 capture can be applied to all fossil fuel and biomass combustion processes. But onlylarge point sources, each emitting quantities in the order of a million tonnes of CO2 per year, canachieve the economies of scale that are needed to make CCS a cost-effective strategy. These point sourcesare electricity production (the main sector where CCS can be applied), manufacturing, and fuel processing.All three are considered in this book. A 500 MW coal-fired power plant with 40% electric efficiency emitsabout 2.5-3.5 Mt CO2 per year, and a similarly sized gas-fired power plant emits about 1-1.5 Mt CO2

per year. Given these quantities, a limited number of projects in certain sectors can have a significantimpact on country level emissions.

CCS in combination with hydrogen production from fossil fuels would result in a fuel that could achievea substantial emission reduction from the transportation sector, a sector where few alternative cost-effective options exist. Such a CO2-free transportation system based on coal or gas with CO2 capture

3. 7.38 MPa equals 73.8 Bar, almost 74 times atmospheric pressure.

037-100 chapter 3 18/11/2004 07:20


would have advantages from a supply security perspective, as it is built on proven large-scale fossil-fuelsupply systems, and would simultaneously reduce dependency on oil.

However, not all intrinsic CCS strategy characteristics are positive. As an add-on technology, CCS wouldincur additional costs and reduce energy efficiency, compared to the same processes without CO2 capture.In principle, other CO2-free energy options could result in lower-cost energy supply. If this is the case, itwould make sense to use these supply options instead of CCS.

From an environmental policy perspective, it is worth bearing in mind that CCS may not always be thecomplete answer to the problem of CO2. Past experience suggests that shifting flows from one mediumto another can create new unforeseen environmental problems.4

So far, there is very little experience with long-term CO2 storage and no proof that storage can be safelyguaranteed over a period of centuries. Moreover while the potential for underground storage is substantial,it is not infinite. In addition, potential storage sites are not evenly distributed around the world. Certainworld regions have substantial underground storage potential while others have none. Therefore, therelevance of CCS will differ by region.

It depends on the reader’s time perspective as to whether CCS should be considered a potential ‘solution’or a ‘transitional strategy’. From a climate change perspective, reducing CO2 emissions by over 75% inkey parts of the energy system in the coming decades is certainly appealing. Such targets may seemoverly ambitious. In fact they are not, if one believes that climate change poses a serious risk.5 Emissionsin developing countries will keep rising as their economies grow. This will put a higher burden on developedcountries both to reduce their own emissions and to help developing countries to reduce theirs.

CO2 Capture Opportunities in the World EnergySystem

The bulk of anthropogenic global CO2 emissions are caused by fossil-fuel energy use. Analysingthe world energy balance can help identify key source categories where CO2 capture and storagecould be applied. CCS is well suited to large stationary point sources, such as power plants, andless appropriate for smaller or dispersed point sources. Therefore, not only the total quantity emittedby a given source category, but also the emission by source, should be considered when assessingthe potential for CO2 capture.

Table 3.1 provides an overview of global energy use in 2000. It shows that total primary energyuse, excluding marine bunkers, amounted to 417 EJ in 2000.6 Of this amount, 125 EJ (30%) wasused in upstream processes (including transformation processes, the energy sector and distribution

4. For example the shift from waste disposal to waste incineration resulted in many countries in a significant increase of dioxin emissions.This problem has been solved by improved filters. This is an example how proper technology development can prevent this shift inenvironmental problems.

5. Recent measurements indicate that rapid climate change in Europe and eastern North America has occurred several times in the past.During these periods, the average temperature dropped by five degrees Celsius over a decade and remained at this level. These changeswere related to variations in the Gulf Stream which transports heat from the tropics to the northern Atlantic. Past changes in the GulfStream were related to changes in salinity. Salinity of the seawater in the North Atlantic has changed dramatically in the past four decades(Dickson et al., 2002). It is thought that such changes are related to the melting of the Greenland glaciers. At some point, this mayaffect the Gulf Stream.

6. 1 EJ (exajoule) = 1000 PJ (petajoule) = 1018 joules.

037-100 chapter 3 18/11/2004 07:20

losses). In the transformation sector, electricity and heat production accounted for almost 83 EJnet energy use (fuel input minus electricity and heat output), or over 100 EJ fossil fuel use, a quarterof total world primary energy use. These energy quantities and their related CO2 emissions showthat the electricity sector is a prime candidate for CO2 capture.

Some 291 EJ (69%) of final energy is used in end-use sectors. Six end-use sectors are defined inthe table: agriculture, residential, services, transport, industry and non-energy use. Industry accountsfor 93 EJ final energy use (32%). Industry-related transformations, energy-sector activities andnon-energy use such as coke ovens, blast furnaces, naphtha steam-cracking and aromatics production,accounted for in the IEA statistics as transformation and energy sector, are in fact industry operations.This industry-related energy use in the other categories amounts to 16 EJ.

The table also shows that final energy use by manufacturing industry amounted to 109 EJ in2000, in the form of electricity and heat. Part of the electricity and most of the heat were derivedfrom industrial combined heat and power (CHP) plants with high efficiency,7 but the bulk (morethan three-quarters) of the electricity was purchased from the grid. Given current efficiencies inelectricity production, industrial demand represented 120-130 EJ primary energy in 2000, about30% of total primary energy use. This brief analysis shows the importance of the manufacturingsector for total global energy use and related environmental impacts, making it a secondimportant category where CO2 capture could be applied.

A third major CO2 emission category is the transport sector (75 EJ). The bulk of transport sectorenergy demand is for road vehicles. The problem with CO2 capture in this sector is the dispersednature of the emissions. CO2 capture technologies for vehicles would be prohibitively expensive.However, switching from petrol to another fuel, such as hydrogen or externally-produced electricityfrom fossil fuels with CO2 capture, could help overcome this problem. It would result in CO2 frompoint sources that could be captured.

The table also shows that the residential, service and agriculture sectors accounted for 40% offinal energy use in 2000. If upstream emissions are included in this figure, the percentage is increasedbecause of the high share of electricity in the sectors’ final energy mix. The dispersed nature of theemissions from fuel combustion in these sectors constitutes a problem similar to that in the transportsector. The optimal solution for these final consumption sectors is not yet clear and is likely to bediverse. Enhancing energy efficiency, increasing electrification and introducing alternative fuelssuch as hydrogen are competing options for CO2-free energy in these market segments.

The relevance of emission categories for CCS will change in the coming decades. IEA projectionssuggest a doubling in electricity demand between 2000 and 2030 (IEA, 2002a). This growth ismuch higher than for primary energy use as a whole, which is projected to grow by 66% duringthe same time period. This will increase the relevance of a CCS strategy for the electricity sector.

Growth in electricity demand is particularly pronounced in developing countries, a factor whichhas significant consequences for the regional urgency of emission reduction. Growth in industrialenergy demand is much lower than in other parts of the energy system. As a consequence, theimportance of the industry sector for CCS will decrease.


7. The IEA fuel questionnaires ask only for an aggregate auto-producer CHP for all end-use sectors. This aggregate is reported under theheading ‘transformation sector’ in the IEA energy statistics. The heat output from CHP plants for own use is translated into a fuel equivalentto produce the same amount of heat in a boiler. This fuel equivalent is excluded from CHP in the transformation sector, and included inthe industrial final end-use. Therefore, a number of other sources must be used in order to analyse industrial CHP.

037-100 chapter 3 18/11/2004 07:20


Table

3.1

Ag

gre

gat

e w

orl

d e

ner

gy

bal

ance

by

sourc

e ca

teg

ory

(2000)

Coal

Biom

ass/

Nat

ural

O

ilN

ucle

arO

ther

El

ectr

icit

yH

eat

Tota

lw

aste

gas

rene

wab

le

(EJ/

yr)

(EJ/

yr)

(EJ/

yr)

(EJ/

yr)

(EJ/

yr)

(EJ/

yr)

(EJ/

yr)

(EJ/

yr)

(EJ/

yr)

Prod

ucti

on

94.6

145

.39

87.8

415

2.96

28.2

811

.54

0.00

0.02

420.

65Fr

om O

ther

Sou

rces

0.

060.

010.

000.

030.

000.

000.

000.

000.

10Im

port

s

17

.16

0.04

22.3

211

6.20

0.00

0.00

1.79

0.00

157.

51Ex

port

s

-1

6.93

-0.0

4-2

2.45

-116

.77

0.00

0.00

-1.7

90.

00-1

57.9

7In

tern

atio

nal M

arin

e Bu

nker

s

0.

000.

000.

00-5

.89

0.00

0.00

0.00

0.00

-5.8

9

Stoc

k Ch

ange

s

2.

700.

050.

52-0

.94

0.00

0.00

0.00

0.00

2.33

Prim

ary

Supp

ly

97

.61

45.4

588

.23

145.

5928

.28

11.5

40.

000.

0241

6.73

Tran

sfer

s0.

000.

000.

000.

520.

000.

000.

000.

000.

52

Stat

istic

al D

iffer

ence

s -0

.69

0.00

-0.4

20.

160.

000.

000.

000.

00-0

.94

Tran

sfor

mat

ion

Sect

or

-7

2.61

-4.1

7-3

1.02

-12.

91-2

8.28

-11.

2755

.25

11.7

7-9

3.26

Elec

tric

ity/

heat

/CH

P8-6

5.35

-2.4

0-2

9.92

-12.

45-2

8.28

-11.

2755

.27

11.7

5-8

2.67

Coke

Ove

ns

-1.6

10.

000.

00-0

.05

0.00

0.00

0.00

0.00

-1.6

6G

as W

orks

-0

.110.

00-0

.08

-0.0

90.

000.

000.

000.

00-0

.28

Blas

t Fu

rnac

es-4

.87

0.00

0.00

-0.0

50.

000.

000.

000.

00-4

.93

Petr

oleu

m R

efin

erie

s0.

00-0

.05

0.00

-1.2

10.

000.

000.

000.

00-1

.27

Liqu

efac

tion

Plan

ts-0

.63

0.00

-0.9

90.

950.

000.

000.

000.

00-0

.67

Char

coal

Pro

duct

ion

0.00

-1.6

70.

000.

000.

000.

000.

000.

00-1

.67

Non

-spec

ified

/ot

hers

-0.0

6-0

.05

-0.0

20.

000.

000.

000.

000.

00-0

.11

Ener

gy S

ecto

r

-2

.08

-0.0

1-8

.16

-8.7

00.

000.

00-5

.06

-1.0

6-2

5.06

Oil

and

Gas

Ext

ract

ion

0.00

0.00

-5.6

9-0

.60

0.00

0.00

-0.4

9-0

.18

-6.9

6Co

ke O

vens

-0

.72

0.00

-0.0

20.

000.

000.

00-0

.01

-0.0

1-0

.77

Blas

t Fu

rnac

es

-0.1

00.

000.

000.

000.

000.

000.

000.

00-0

.10

BCB

Plan

ts-0

.01

0.00

0.00

0.00

0.00

0.00

0.00

0.00

-0.0

1Pe

trol

eum

Ref

iner

ies

-0.0

10.

00-1

.55

-7.8

20.

000.

00-0

.64

-0.5

9-1

0.62

LNG

Pla

nts

0.00

0.00

-0.6

40.

000.

000.

000.

000.

00-0

.65

Ow

n U

se E

lec/

heat

-0

.56

0.00

-0.0

4-0

.10

0.00

0.00

-3.17

-0.0

9-3

.97

Non

-spec

ified

/ot

hers

-0.6

80.

00-0

.19

-0.0

60.

000.

00-0

.55

-0.0

9-1

.57

037-100 chapter 3 18/11/2004 07:20


Dis

trib

utio

n Lo

sses

-0

.13

0.00

-0.8

7-0

.20

0.00

-0.0

1-4

.80

-0.8

1-6

.81

Fina

l Con

sum

ptio

n

22.1

041

.27

47.7

612

4.46

0.00

0.27

45.4

09.

9329

1.19

Indu

stry

Sec

tor

6.53

6.73

22.0

424

.86

0.00

0.02

19.1

03.

9893

.27

Iron

and

Stee

l5.

540.

202.

160.

730.

000.

002.

240.

4011

.28

Chem

ical

1.99

0.20

8.95

14.3

00.

000.

003.

041.

2929

.77

Non

-Fer

rous

Met

als

0.47

0.00

0.84

0.42

0.00

0.00

1.96

0.26

3.95

Non

-Met

allic

Min

eral

s3.

760.

151.

671.

500.

000.

001.

010.

128.

21Tr

ansp

ort

Equi

pmen

t0.

140.

000.

400.

190.

000.

000.

590.

041.

36M

achi

nery

0.37

0.00

0.99

0.42

0.00

0.00

1.67

0.45

3.90

Min

ing

and

Qua

rryi

ng0.

160.

000.

190.

460.

000.

000.

640.

021.

47Fo

od a

nd T

obac

co0.

660.

861.

350.

810.

000.

001.

090.

355.

12Pa

per,

Pulp

and

Prin

t0.

562.

101.

230.

650.

000.

011.

610.

126.

27W

ood/

Woo

d Pr

oduc

ts0.

060.

450.

120.

150.

000.

000.

320.

281.

38Co

nstr

uctio

n0.

160.

000.

110.

610.

000.

000.

170.

071.

13Te

xtile

and

Lea

ther

0.34

0.01

0.39

0.56

0.00

0.00

0.69

0.15

2.14

Non

-spec

ified

2.31

2.76

3.63

4.06

0.00

0.01

4.09

0.44

17.2

9Tr

ansp

ort

Sect

or0.

250.

352.

2571

.23

0.00

0.00

0.80

0.00

74.8

7In

tern

atio

nal A

ir0.

000.

000.

004.

970.

000.

000.

000.

004.

97D

omes

tic A

ir0.

000.

000.

004.

380.

000.

000.

000.

004.

38Ro

ad0.

000.

350.

1258

.91

0.00

0.00

0.00

0.00

59.3

8Ra

il0.

240.

000.

001.

370.

000.

000.

590.

002.

20Pi

pelin

e Tr

ansp

ort

0.00

0.00

2.12

0.00

0.00

0.00

0.11

0.00

2.23

Inte

rnal

Nav

igat

ion

0.00

0.00

0.00

1.43

0.00

0.00

0.00

0.00

1.43

Non

-spec

ified

0.00

0.00

0.01

0.17

0.00

0.00

0.09

0.00

0.27

Agr

icul

ture

0.51

0.25

0.27

4.27

0.00

0.00

1.27

0.23

6.80

Com

mer

ce/

Serv

ices

0.50

0.22

6.04

4.68

0.00

0.01

10.5

40.

9722

.96

Resi

dent

ial

3.53

32.7

115

.66

9.95

0.00

0.21

12.8

44.

4079

.30

Non

-spe

cifie

d0.

381.

001.

512.

020.

000.

010.

840.

356.

12N

on-E

nerg

y U

se

0.41

0.00

0.00

7.46

0.00

0.00

0.00

0.00

7.86

8. T

his

incl

udes

CH

P in

all

sect

ors,

but i

t exc

lude

s in

dust

rial s

t eam

pro

duct

ion

in C

HP

plan

ts fo

r ow

n us

e (w

hich

is a

ccou

nted

for a

s in

dust

rial f

uel u

se).

BCB

= Br

own

Coal

Briq

uett

es; L

NG

= L

ique

fied

Nat

ural

Gas

. Nuc

lear

ele

ctric

ity e

xpre

ssed

in p

rimar

y nu

clea

r ene

rgy

equi

vale

nts

assu

min

g 33

% e

ffic

ienc

y. O

ther

rene

wab

les

= hy

dro,

win

d an

d so

l ar e

lect

ricity

.10

0% e

ffic

ienc

y, ge

othe

rmal

10%

eff

icie

ncy.

Sour

ce: B

ased

on

IEA

ene

rgy

stat

istic

s.

037-100 chapter 3 18/11/2004 07:20


Table

3.2

Glo

bal

CO

2em

issi

on

s by

sourc

e ca

teg

ory

(2000)

Coal

Biom

ass/

Nat

ural

O

ilIn

orga

nic

Tota

lw

aste

gas

rene

wab

le

(Gt

CO2/

yr)

(Gt

CO2/

yr)

(Gt

CO2/

yr)

(Gt

CO2/

yr)

(Gt

CO2/

yr)

(Gt

CO2/

yr)

Tota

l 9.

085.

224.

8910

.21

0.77

30.1

3of

whi

ch:

Tran

sfor

mat

ion

Sect

or

6.

220.

491.

740.

949.

36

Elec

tricit

y an

d he

at p

lant

s6.

140.

281.

680.

919.

00

Gas

Wor

ks0.

010.

000.

000.

010.

02

Petr

oleu

m R

efin

erie

s0.

000.

010.

000.

090.

09

Liqu

efac

tion

Plan

ts0.

060.

000.

06-0

.07

0.05

Char

coal

Pro

duct

ion

0.00

0.19

0.00

0.00

0.19

Non

-spec

ified

/ot

hers

0.01

0.01

0.00

0.00

0.01

Ener

gy S

ecto

r

0.

110.

000.

460.

621.

20

Oil

and

Gas

Ext

ract

ion

0.00

0.00

0.32

0.04

0.36

Petr

oleu

m R

efin

erie

s0.

000.

000.

090.

570.

66

LNG

Pla

nts

0.00

0.00

0.04

0.00

0.04

Ow

n-us

e el

ectri

city

and

hea

t0.

050.

000.

000.

010.

06

Non

-spec

ified

/ot

hers

0.06

0.00

0.01

0.00

0.08

Fina

l Con

sum

ptio

n +

bunk

ers

+co

ke o

vens

/bl

ast

furn

aces

2.75

4.73

2.69

8.65

0.77

19.5

7

Indu

stry

Sec

tor

2.26

0.77

1.24

1.92

0.77

6.97

Iron

and

Stee

l inc

l. CO

&BF

91.

210.

020.

120.

050.

071.

48

Chem

ical i

ncl.

non-

ener

gy0.

210.

020.

501.

160.

001.

89N

on-F

erro

us M

etal

s0.

040.

000.

050.

030.

12

9. C

O&

BF =

Cok

e O

vens

and

Bla

st F

urna

ces

037-100 chapter 3 18/11/2004 07:20


Non

-Met

allic

Min

eral

s0.

350.

020.

090.

110.

701.

27

Tran

spor

t Eq

uipm

ent

0.01

0.00

0.02

0.01

0.05

Mac

hine

ry0.

030.

000.

060.

030.

12

Min

ing

and

Qua

rryi

ng0.

020.

000.

010.

030.

06

Food

and

Tob

acco

0.06

0.10

0.08

0.06

0.30

Pape

r, Pu

lp a

nd P

rint

0.05

0.24

0.07

0.05

0.41

Woo

d/ W

ood

Prod

ucts

0.01

0.05

0.01

0.01

0.07

Cons

truc

tion

0.02

0.00

0.01

0.04

0.07

Text

ile a

nd L

eath

er0.

030.

000.

020.

040.

10

Non

-spec

ified

0.

220.

320.

200.

301.

03

Tran

spor

t Se

ctor

0.02

0.04

0.13

5.20

5.39

Inte

rnat

iona

l Mar

ine

Bunk

ers

0.00

0.00

0.00

0.43

0.43

Agr

icul

ture

0.05

0.03

0.02

0.31

0.40

Com

mer

ce/

Serv

ices

0.05

0.03

0.34

0.34

0.75

Resi

dent

ial

0.33

3.75

0.88

0.73

5.68

Non

-spe

cifie

d0.

040.

110.

080.

150.

38

Not

e: In

clud

es e

nerg

y an

d in

orga

nic

CO2

emis

sion

s; ex

clud

es d

efor

esta

tion.

No

corre

ctio

n fo

r car

bon

sto r

age

in s

ynth

etic

org

anic

che

mic

als.

037-100 chapter 3 18/11/2004 07:20


Table 3.2 shows CO2 emissions by source category. The emissions were calculated based on theyear 2000 energy balance shown in Table 3.1. The energy use is multiplied by the carbon contentper unit of energy.10

Emissions from limestone dissociation (mainly in cement, iron and glass production) have beenadded. The table shows that total global emissions amounted to almost 30 Gt CO2 in 2000, ofwhich 5.2 Gt CO2 were from biomass combustion. While these emissions are usually not taken intoaccount in CO2 emission calculations,11 there is no physical difference between CO2 from fossil fuelcombustion and CO2 from biomass combustion. CO2 capture could be applied to both cases.

Oil is the most important fuel from a CO2 emissions perspective, closely followed by coal. Biomassand natural-gas related emissions are on par. From a sector perspective, electricity and heat plantsconstitute the single most important emissions source, followed by the industry sector.

CO2 Capture in the Electricity Sector

This section discusses the following three main groups of technologies which can be used to captureCO2 from power plants. Such options can be combined and integrated with various process designs:

● CO2 capture from flue gas;

● Fuel reforming into hydrogen and CO2, followed by capture from the concentrated and pressurizedgas;

● The use of oxygen for combustion, which results in a concentrated CO2 flue gas.

CO2 capture options for power plantsThe following assessment of CO2 capture options for power plants is split into CO2 capture for newplants and CO2 capture for existing plants. The assessment of capture technology for new plantsis further divided into existing capture technologies, emerging capture technologies, and an overviewof technology efficiencies and cost.

Existing capture technologies for new power plants: post-combustion chemicaland physical absorption

CO2 capture is already widely applied in industrial manufacturing processes, refining and gasprocessing. These capture technologies can also be applied to power plants. In the 1980s, CO2

capture from gas-fired boiler flue gases was applied commercially in order to produce CO2 forenhanced oil recovery (EOR) projects (Chapel et al., 1999). These processes were commercially viableat a price of 19-38 USD/t CO2. However, the plants were closed when the oil price collapsed.

Existing CO2 capture systems are either based on chemical absorption, in combination with heatinduced CO2 recovery (using solvents such as MonoEthanolAmine MEA), or physical absorption

10. This is a crude approach, especially for transformation processes. For example, the carbon content of coal equals 94 kg CO2/GJ,while the emission coefficient of blast furnace gas is 242 kg CO2/GJ. A significant share of blast furnace gas is sold by the iron andsteel industry to electricity producers. This CO2 is emitted by power plants using blast furnace gas. This is not reflected in Table 3.2. Infuture, CO2 may be removed from the blast furnace gas before it is delivered to the power plants, so one could argue that this is a morerealistic allocation for an analysis of emission reduction potentials. According to IEA statistics, emissions from fossil fuel use amountedto 23.4 Gt in 2000. This includes a correction for carbon storage in synthetic organic chemicals of almost 1 Gt CO2.

11. This is correct if the carbon in plants and soil recovers to its original level. Usually this will be the case for plantations. In fact suchplantations may store CO2 from the atmosphere, as they increase the soil carbon content.

037-100 chapter 3 18/11/2004 07:20


Evaluating the cost of CCS: different methods yield different results

CO2 capture and pressurization (necessary for transport and injection) increases energyuse which results in additional emissions that must be taken into account when evaluatingthe impact and cost-efficiency of CCS (Freund, 2003). The CO2 capture cost and CO2avoidance cost require two different evaluation methods. For power plants, capture costcan be translated into avoidance cost based on the equation:

Cost(avoided) = Cost(captured) x CE/[effnew/effold – (1-CE)]

where:

effnew and effold is the efficiency of the power plants with and without CO2 capture,respectively, and CE is the fraction of CO2 that is captured. Cost expressed per t of CO2avoided is higher than costs expressed per t of CO2 captured. For example, in case ofeffnew is 31% and effold is 43%, and CE is 0.85, the correction factor is 1.48. The correctionfactor declines to 1.20-1.25 for energy efficient emerging CCS technologies.

Only CCS cost expressed per t of CO2 avoided allow for comparison with other CO2 abatementmeasures in terms of cost of environmental effects that have been achieved. Full economicanalysis of technology options requires, however, introduction of another parameter thatrelates cost to the technology output. For power generation sector this would mean usingcost of CO2-free electricity. Using such cost parameter entails making additional assumptionsconcerning power plant capital cost, discount rates, plant’s lifespan and others.

Comparison of cost expressed per unit of output (e.g., per kWh of CO2-free electricityproduced) can yield different results than a comparison of cost per tonne of CO2 capturedor CO2 avoided. A typical example is that the per tonne cost for CO2 (captured or avoided)will be lower for a coal-fired power plant than for a gas-fired power plant, while the electricitysupply cost may be lower for the gas-fired plant with CO2 capture. All three CCS costparameters (USD/kWh of CO2-free electricity, USD/t of CO2 avoided, USD/t of CO2captured) are being used throughout the book. Modelling analysis, however, is based solelyon the “output” cost parameter in USD/kWh of CO2-free electricity.

CCS for a coal-fired power plant will reduce emissions significantly, compared to thesame power plant without CO2 capture. However, comparing an identical plant with andwithout CO2 capture may not adequately reflect the real emission impact in the case ofa green-field investment decision. A coal-fired power plant with CCS does not reduceemissions compared to a hydropower or nuclear plant. Therefore, the choice of a referenceprocess is crucial for estimating CO2 avoidance costs.

In a marginal costing approach, the reference plant is the plant with the highest supply costsin the base case without CO2 policies, i.e., the plant that determines the product price in anideal market. The emissions of this plant may be high or low, depending on the energy resourceendowment and economic structure of a region. For many OECD countries, a gas-fired combinedcycle power plant would be the marginal producer to which a coal-fired power plant withCO2 capture should be compared. This reduces the CO2 benefits by a half or even two-thirds.

037-100 chapter 3 18/11/2004 07:20


Upstream emissions of CO2 also need to be taken into account when considering coal orgas life cycles. The characteristics of the specific supply chain must be accounted for sinceglobal averages make no sense. Depending on the supply chain, upstream CO2 emissionscan amount to between 0% and 20% of power plant emissions. Upstream emissions maydecline in the future because of technological progress and reduced leakages, althoughthis trend may be balanced by increasing transportation distances and a shift from pipelineuse to shipped LNG, driven by resource exhaustion. The net effect will likely be a slightdecline in emissions. In the model analysis outlined later in this book, upstream emissionsare accounted for and based on region-specific energy supply structures.

(using solvents such as dimethylether of polyethylene glycol, so-called Selexol), in combinationwith pressure-induced CO2 recovery. A range of solvents is being studied as outlined in Table 3.3.

The chemical absorption process is inherently energy inefficient due to the energy needed to breakthe strong bonding of the solvent and CO2. As a result, new chemical absorbents, such as so-calledsterically-hindered amines, are being investigated. The bonding strength between the solvent andCO2 is lower than for MEA. As a consequence, less energy is needed to release the CO2 from thesolvent. Steam consumption for the latest chemical absorption systems is on average about 1.5 tonnesof low pressure steam per tonne of CO2 recovered (3.2 GJ/t) for a boiler system with 90% recovery(it is slightly higher for higher recovery rates; Mimura et al. 2002). The recovery energy declinesfrom 3.4 to 2.9 GJ/t for CO2 concentrations increasing from 3% to 14%. The extremes representthe conditions for natural gas turbines and coal-fired steam cycles.

Physical absorption is based on the weak binding of CO2 and the solvent. Binding takes place athigh pressure with the CO2 released when the pressure is reduced. The only energy needed for CO2

capture is the electricity for gas pressurization. The amount of energy per tonne of CO2 is proportionalto the inverse of the CO2 concentration in the gas: twice as much energy is needed if the CO2

concentration in the gas stream is halved. Chemical absorption is the preferred method at lowCO2 concentrations (lower than 10%, such as flue gases from gas-fired power plants) becauseits energy use is not particularly sensitive to low concentrations and low partial pressures ofCO2. Physical absorption is the preferred method at higher CO2 concentrations (higher than15%) and at higher partial pressures.

The flue gases from a gas-fired combined cycle power plant contain between 3% and 4% of CO2

and those from a conventional coal-fired power plant between 13% and 14% of CO2. These relativelylow concentrations are a result of using air for the combustion process. From a combustion perspective,only oxygen is needed. However, air contains about 80% nitrogen and 20% oxygen. The nitrogendilutes the CO2 in the flue gas. Capture is easier at higher CO2 concentrations. Higher CO2

concentrations can be achieved in two ways. The first is by using pure oxygen instead of air.The second is by converting the fuel gas into CO2 and H2, and pre-combustion CO2 removal,before the air is added for combustion.

Emerging capture technologies for new power plants: pre-combustion captureand combustion using pure oxygen

Gasification technology, shift reactors, air separation, hydrogen separation and hydrogen turbinesplay a crucial role in the pre-combustion removal of CO2 (Dijkstra and Jansen, 2003). ExistingGeneral Electric F-class turbines can accept gas containing 45% H2. However, further developmentof gas turbines is required before pure hydrogen can be used to generate electricity. For example,

037-100 chapter 3 18/11/2004 07:20

combustion control for hydrogen-fuelled gas turbines requires better control of process parametersthan for natural gas turbines. NOx emissions also need to be reduced further to acceptable levels,without excessive water/steam injection (Roekke, 2003).

The efficiency of pre-combustion CO2 separation in natural gas reforming, including the use of membranes,is expected to be slightly higher than for current post-combustion absorption systems (NorwegianPetroleum Directorate, 2002). Compared to post-combustion absorption membrane systems, the efficiencygains of pre-combustion separation systems for natural gas are marginal. The uncertainty of technicaldata is higher than the projected efficiency gains. For coal, the pre-combustion removal of CO2 isgenerally considered to be advantageous compared to post-combustion removal.12

Oxyfueling is another promising strategy for CO2 capture. By using oxygen instead of air, a relativelypure CO2 stream is created during combustion. Oxyfueling can be applied to steam cycles and gasturbines but it requires an air separation unit. In the case of a gas turbine, a process redesign isneeded in order to maintain an acceptable temperature in the gas turbine. One option is to recycle


Table 3.3

Commercial CO2 scrubbing solvents used in industry

Solvent name Solvent type Process conditions

Rectisol Methanol -10/-70°C, >2 MPaPurisol n-2-methyl-2-pyrolidone -20/+40°C, >2 MPa

Physical solvents Selexol Dimethyl ethers of polyethyleneglycol -40°C, 2-3 MPaFluor solvent Propylene carbonate Below ambient temperatures,

3.1-6.9 MPa

MEA 2,5 n monoethanolamine and inhibitors 40°C, ambient-intermediatepressures

Amine guard 5n monoethanolamine and inhibitors 40°C, ambient-intermediatepressures

Econamine 6n diglycolamine 80-120°C, 6.3 MPa

Chemical solventsADIP 2-4n diisopropanolamine 2n

methyldiethanolamine 35-40°C, >0.1 MPaMDEA 2n methyldiethanolamine

Flexsorb,KS-1,KS-2,KS-3 Hindered amine

Benfield and versions Potassium carbonate & catalysts. Lurgi & Catacarb processes with 70-120°C, 2.2-7 MPaarsenic trioxide

Sulfinol-D, Sulfinol-M Mixture of DIPA or MDEA, water and tertahydrothiopene (DIPAM) >0.5 MPaor diethylaminePhysical/

chemical solventsAmisol Mixture of methanol and MEA, DEA,

diisopropylamine (DIPAM) 5/40°C, >1 MPaor diethylamine

Source: Gupta et al., 2003.

12. Note that some studies suggest that post-combustion capture would be preferable, e.g., Canadian Clean Power Coalition, 2004. Suchstudies are usually based on the current state of technology, and they do not account for long-term cost reduction potentials. Advancesin different power plant technologies pose a source of uncertainty that can affect the choice of the optimal CCS technology. Also thequality of the coal (hard coal or lignite) can influence the results to some extent. See also the technology roadmap in Chapter 8.

037-100 chapter 3 18/11/2004 07:20

CO2 to cool the turbine (for example the so-called MATIANT cycle). A second option is to use amixture of steam and exhaust as a gas turbine working medium (so-called Graz cycle; Gupta et al.,2003). The MATIANT cycle would require the development of new turbines since retrofitting existingturbines would not be feasible from a technical perspective.

Some suggest that the efficiency of gas and even coal oxyfuel systems will be lower than for post-combustion absorption due to the high oxygen requirements and the energy use for oxygen production(Canadian Clean Power Coalition, 2004). This does not account for possible substantial improvementsin oxygen production (see box). While there is some debate as to whether oxyfueling hasadvantages for natural gas combined cycles and coal-fired steam cycles, oxygen blown gasifierswith pre-combustion CO2 removal in combination with hydrogen turbines will be essential forcoal-fired IGCCs.

Chemical looping is another oxygen supply concept worth mentioning here. The concept is basedon the use of a metal/metal oxide system to provide a reversible chemical reaction for oxygensupply. In one reactor the metal reacts with air to produce a metal oxide; in another reactor, themetal oxide reacts with the fuel to produce syngas and metal (so-called flameless combustion).


The role of membranes in CO2 capture

Gas separation membranes are likely to play a key role in CO2 capture systems in the future.Their energy efficiency can be higher than for absorption separation systems, as a limitedpressure drop across the membrane is sufficient to achieve separation. Their modular designalso allows their use in combination with small-scale modular fuel cells, a power plantconcept for the future. While membranes are widely applied for gas separation, they arenot yet applied on a power plant scale.

The disadvantage of membrane separation systems for CO2 capture is that their separationefficiency is relatively poor. Only a fraction of the CO2 is recovered and the purity of theCO2 is relatively low. For example, a study on membranes for fuel gas CO2 separationfrom IGCC plant suggests that only multi-stage membrane systems can meet the necessarypurity criteria. For IGCCs, ceramic membranes are preferable over polymer membranes, asthey can operate at higher temperatures, thus reducing the need for cooling of fuel gases.Polymer membranes, however, are more developed than ceramic membranes and canachieve much higher CO2 recoveries, around 57% compared to 7%.

At present, using polymer membrane separation systems would increase investment costsfrom 1,263 USD per kW to 5,700 USD/kW (Kaldis et al., 2003). Clearly, further costreductions are needed as are improvements in separation efficiency for ceramic membranes.It is possible to combine membrane separation systems with absorption systems in orderto achieve a higher efficiency of CO2 separation. This type of combined separation systemhas been considered in the analysis.

In addition to gas separation membranes, gas absorption membranes offer high capturepotential. These membranes work as contacting devices between gas and liquid flow. Theirfunction is to increase the contact area, thus reducing the size of the scrubbing equipment(McKee, 2002). These absorption membranes increase energy efficiency and reduce capturecost, but to a lesser extent than the speculative, future gas separation membranes.

037-100 chapter 3 18/11/2004 07:20

Metal and metal oxide are transported from one reactor to the other. Such a system avoids energyintensive air separation for pure oxygen supply. Studies suggest that the electricity productioncost for a coal-fired circulating fluidized bed with chemical looping and CO2 capture would belower than for IGCC (Nsakala et al., 2003). Studies suggest up to 54% efficiency for a gas-fired power plant with chemical looping and CO2 capture, including CO2 pressurization (Brandvolland Bolland, 2002).

The concept of chemical looping for oxygen supply has been around for more than 25 years, buthas only been applied in the laboratory, not on a commercial scale. In the past, similar processdesigns based on particle transfer have experienced plugging and abrasion problems. Metal oxidematerials are needed that withstand chemical cycling and are resistant to physical and chemicaldegradation caused by impurities from fuel combustion (Gupta et al., 2003). The system alsoneeds to be proven on a pilot plant scale. CO2 capture based on chemical looping – for coal andfor gas-fired electricity production – should be considered a speculative technology.

For a coal fired IGCC the electricity used for oxygen production amounts to 10% of the IGCCelectricity production. This may be halved in the future (see the box above). The energy needed forCO2 pressurization represents some 8% of the electricity output. Gasification efficiency is somewherebetween 75% and 90% and depends on the gas cleaning technology. Low temperature gas cleaningis a proven technology but it results in energy losses compared to future high temperature gascleaning, which is not yet proven on a commercial scale. The energy requirements for the shiftreactor amount to 4% of the syngas LHV (Mathieu, 2003). This assessment does not account forgas turbine efficiency variations due to differences in the gas composition.


The importance of new air separation technologies

The efficiency of power plants and CO2 capture systems using oxygen depends criticallyon the energy required for oxygen production. At present, large-scale oxygen production isbased on cryogenic air separation with plants reaching capacities of up to 3000 t of oxygenper day. Energy consumption required for this has declined to around 0.3 kWh/Nm3 lowpressure oxygen (210 kWh/t oxygen or 0.77 GJel/t oxygen). A further reduction to0.28 kWh/Nm3 is projected for 2010 (a 6.7% energy efficiency improvement).

More complex processes at higher pressures may reduce power consumption further andresult in capital cost savings (Castle, 2002). Vacuum Pressure Swing adsorption is analternative for medium size plants up to 250-350 t of oxygen per day. A typical 250 MWIGCC needs 2,000 t of oxygen per day. Ion transport membrane systems, based on inorganicoxide ceramic materials, could be used to provide oxygen for IGCCs. What is not clear iswhether this technology, which is still under development, will be economical when scaledup for use in power plants (Smith and Klosek, 2001). If membrane systems do succeed,the energy requirement for air separation may be reduced to 147 kWh/t oxygen (Steinand Foster, 2001). This would represent a 51% energy efficiency improvement, comparedto the current cryogenic oxygen separation technology.

For an oxygen-blown IGCC this would imply an electric efficiency increase of 1-2 percentagepoints (2-5% in relative terms). At the same time, the costs of oxygen production are reducedby 35% and the investment costs for IGCC reduced by 75 USD/kW. These figures suggestthat new air separation systems would enhance the prospects of oxyfuelling- based CO2capture strategies significantly.

037-100 chapter 3 18/11/2004 07:20

An IGCC with CO2 capture can be considered as a gas-based combined cycle where some processeshave been added. The additions are coal gasification, oxygen production, shift reactor and CO2

separation. Assuming a 60% efficiency for a gas-fired combined cycle without CCS (the total efficiencyfor the gas turbine and the steam cycle), a coal-fired IGCC with CCS can achieve between 36 and 43%efficiency. This back-of-envelope calculation shows where the main losses occur. Efficiency gains can beachieved in several ways. For example, efficiency in electricity generation can be increased by addingfuel cells; increasing gasification efficiency is achievable by using high-temperature gas cleaning; reducingthe energy required for air separation unit can be done by using membrane separation processes.

IGCC designs are based on the use of oxygen and steam for coal gasification at high pressure,conditions that make the plant well suited for fuel gas CO2 removal. However, this is balanced bythe comparatively high cost of IGCCs and the relatively immature state of this technology. Estimatesof future IGCC investment costs vary considerably. The variations can be attributed to a range ofissues. For example, when high availability is required, a spare gasifier is needed, which increasesinvestment costs by between 150 and 200 USD/kW.

At present, only so-called F class gas turbines are available. In future, H class gas turbines may becomeavailable. These would increase electric efficiency by 1.3 to 3.4 percentage points. Since higher gasturbine efficiencies imply smaller gasifiers, this would in turn reduce investment costs per kW by 10-20%. Three types of gasifiers exist: one-stage slurry fed such as the Texaco gasifier, two-stage slurryfed gasifiers such as the E-Gas gasifier, and dry fed systems such as the Shell gasifier. The efficiency ofthe dry fed systems is significantly higher, but so is the cost (IEA GHG, 2003). Furthermore, the typeof coal used influences cost. For example, building a lignite-fired IGCC costs 400 USD/kW more thanusing a hard coal-fired IGCC (Breton and Amick, 2002). Some IGCC costing studies account for contingencycost, others do not. This makes a difference of 100 to 200 USD/kW. Furthermore, additional CO2 capturecosts differ depending on IGCC design, ranging from 350 USD/kW for Texaco quench-type gasifiersto 550 USD/kW for Shell and E-gas designs.


New capture technologies

In the quest for more energy efficient and less costly capture technologies, new technologyconcepts are being investigated. One concept is based on adsorption of CO2 to solids.Temperature swing adsorption, pressure swing adsorption and electric swing adsorptionare examples of this. However, adsorption is not yet considered attractive for large-scaleseparation of CO2 from flue gas because the capacity and CO2 selectivity of availableadsorbents are low (Gupta et al., 2003).

Another strategy which has been proposed is based on the CaO/CaCO3 system in whichlime (CaO) is added to the combustion process in a fluidized bed. The CaO reacts withthe CO2 produced from fuel combustion. The CaCO3 formed is recycled to another reactor,where it is calcined (dissociation at temperatures above 850 °C). The challenge is to identifyprocess conditions where the CaO remains active. So far, such conditions have not beenfound (Salvador et al., 2003).

Given the low likelihood of these capture systems and their limited benefits compared toother more likely capture systems such as chemical and physical absorption, they havenot been considered in this book’s analysis.

037-100 chapter 3 18/11/2004 07:20

While some experts question the attractiveness of IGCC compared to coal-fired steam cycles in aCO2-unconstrained world, the attractiveness of IGCC increases compared to conventional steamcycles, when CO2 capture is required. This is true for both hard coal and lignite-based electricityproduction systems (Ewers et al., 2003). However, it remains to be seen if this advantage is sufficientto tilt the balance in favour of IGCC.

In the long run, power plants including fuel cells may allow even higher efficiencies than today’s gasturbines and steam cycles. Engineering studies suggest that certain designs would be well suited toCO2 capture, as fuel cells need hydrogen as a fuel. For the time being, such systems are speculative,because these fuel cells have not yet been proven on a commercial MW-scale. Such power plant systems,including solid oxide fuel cells (SOFCs), have been considered in the analysis (Dijkstra and Jansen, 2003).

Overview of capture technology efficiencies and cost for new power plants

Table 3.4 provides an overview of the characteristics of power plants with CO2 capture. All cases havebeen assessed based on a product CO2 flow at 100 bar meaning that CO2 compression is included inthe efficiency losses. The efficiency loss due to CO2 capture ranges from 12 percentage points for existingcoal-fired power plants to 4 percentage points for future designs with fuel cells.

With regard to electricity costs, the gas-based systems with CO2 capture seem cheapest. However, thisresult depends on local fuel prices and discount rates. Moreover, differences between coal and gas-fired systems are relatively small. The figures suggest a prospective electricity cost price increase of 1-2 US cents per kWh. This is an important increase, compared to production costs (+25-50%, seeTable 3.4). However, electricity consumer prices are considerably higher than producer costs. The averageelectricity price was 10.6 US cents/kWh for households in OECD Member countries in 2000. The increaseamounts to 10-20% of the consumer price.

CO2 capture is energy intensive and results in increased coal and gas use for electricity production.The increase ranges from 39% for current designs to 6% for advanced designs (Table 3.4). This isa substantial increase with impacts on global coal and gas markets especially if CCS is widely applied.However, substitution effects resulting in fuel demand reductions can potentially be more substantialthan fuel demand increases because of CCS use, resulting in a net fuel demand decrease on a globalscale. Also CO2 policies will result in a shift to higher efficiency power plants.

Biomass-fired power plants with CO2 capture constitute a special option. Because renewable biomassis a CO2-neutral energy carrier, combining biomass-fired power plants with CO2 capture results in a netremoval of CO2 from the atmosphere (Möllersten et al., 2003; Möllersten et al., 2004). However, theproblem with biomass is generally that the scale of operations is much smaller than for fossil-fuelledpower plants. A typical biomass IGCC would have a capacity of 25 to 50 MW, compared to a coal-firedIGCC with a capacity of 500 to 1000 MW. As a consequence, investment costs per kilowatt are twiceas high for biomass. Also, biomass is considerably more expensive than coal in most world regions. Ina CO2-constrained world, the removal of CO2 from the atmosphere may offset these disadvantages.Moreover, certain industrial biomass conversion processes, such as black liquor gasifiers in pulp production,generate CO2 in quantities of a similar order of magnitude as power plants (see section on industrialprocesses). Finally, a certain amount of biomass can be co-combusted in coal-fired plants.

Existing capture technologies are all characterized by relatively high costs per tonne of CO2 and lowenergy efficiencies, compared to what society today is willing to pay for CO2 emission mitigation. Energyefficiency losses due to CO2 capture and pressurization play a key role. R&D is aiming for new technologieswith higher efficiencies. Such developments are deemed critical for successful large-scale introductionof CCS (Klara, 2003). However, as the complexity of the designs increases so do capital costs. Systemsintegration problems also tend to increase. A number of new conceptual designs seem attractive, but


037-100 chapter 3 18/11/2004 07:20


Table

3.4

Ch

arac

teri

stic

s of

pow

er p

lan

ts w

ith

an

d w

ith

out

CO

2ca

ptu

re

Fuel

, tec

hnol

ogy

Star

ting

INV

FIX

Eff

Eff.

Add

. Ca

pt.

Capt

. cos

tsEl

. cos

tsA

dd. e

l. co

sts

year

(USD

/kW

)(U

SD/

kW.y

r)(%

)lo

ss (

%)

fuel

(%

)ef

f. (%

)(U

SD/

t CO

2)(M

ils/

kWh)

(Mils

/kW

h)

Like

ly t

echn

olog

ies

No

CO2

capt

ure

Coal

, ste

am c

ycle

2010

1,07

523

4329

.1

Coal

, ste

am c

ycle

20

201,

025

3144

29.2

Coal

, USC

ste

am c

ycle

2020

1,26

030

5031

.5

Coal

, IG

CC20

101,

455

5746

37.4

Coal

, IG

CC12

2020

1,26

035

4633

.0

Gas

, CC

2005

400

1456

26.1

Gas

, CC

2015

400

1459

25.2

Blac

k liq

uor,

IGCC

2020

1,30

050

2823

.5

Biom

ass,

IGCC

2020

2,40

050

4074

.6

With

CO

2ca

ptur

e

Coal

, ste

am c

ycle

, CA

2010

1,85

080

31-1

239

8524

51.0

21.9

Coal

, ste

am c

ycle

, m

embr

anes

+C

A20

201,

720

7536

-822

8521

46.3

17.1

Coal

, USC

ste

am c

ycle

, m

embr

anes

+C

A20

301,

675

4542

-819

9517

49.0

17.5

Coal

, IG

CC, S

elex

ol20

102,

100

9038

-821

8520

52.3

14.9

Coal

, IG

CC, S

elex

ol20

201,

635

5040

-615

8511

41.0

8.0

Gas

, CC,

CA

2010

800

2947

-919

8529

36.8

10.7

Gas

, CC,

Sel

exol

/O

xF20

2080

033

51-8

1685

2534

.89.

6

Blac

k liq

uor,

IGCC

2020

1,62

050

25-3

1285

427

.94.

4

Biom

ass,

IGCC

2025

3,00

010

033

-721

8523

96.1

21.5

037-100 chapter 3 18/11/2004 07:20


Spec

ulat

ive

tech

nolo

gies

No

CO2

capt

ure

Coal

, IG

CC &

SO

FC20

301,

800

7560

41.3

Gas

, CC

& S

OFC

2025

800

4070

30.6

With

CO

2ca

ptur

e

Coal

, CFB

,

Chem

ical

loop

ing

2020

1,40

045

39-5

1385

1438

.214

.7

Gas

, CC,

Chem

ical

loop

ing

2025

900

2556

-47

8533

34.5

9.3

Coal

, IG

CC &

SO

FC20

352,

100

100

56-4

710

013

49.0

7.7

Gas

, CC

& S

OFC

2030

1,20

060

66-4

610

028

39.2

8.6

Not

e: T

he a

bove

com

paris

on is

bas

ed o

n a

10%

dis

coun

t ra

te a

nd a

30-

year

pro

cess

life

span

. The

inve

stm

ent

cost

exc

lude

s in

tere

s t d

urin

g co

nstr

uctio

n an

d ot

her o

wne

rs’ c

osts

, whi

ch c

ould

add

5-4

0% t

o th

eov

erni

ght c

onst

ruct

ion

cost

. Thi

s ap

proa

ch h

as b

een

appl

ied

to a

ll te

chno

logi

es th

at a

re c

ompa

red

in th

is s

tudy

. Coa

l pric

e =

1.5

USD

/G

J; g a

s pr

ice

= 3

USD

/G

J. CO

2pr

oduc

t in

a su

perc

ritic

al s

tate

at 1

00 b

ar.

CO2

tran

spor

tatio

n an

d st

orag

e is

not

incl

uded

. Cap

ture

cos

ts a

re c

ompa

red

to t

he s

ame

pow

er p

lant

with

out

capt

ure.

Cap

ture

cos

ts a

r e e

xpre

ssed

per

ton

ne o

f CO

2ca

ptur

ed –

see

box

on

eval

uatin

g th

e co

stof

CCS

in t

his

chap

ter

for

conv

ersi

on f

acto

rs t

o co

st p

er t

onne

of

CO2

avoi

ded.

CA

= C

hem

ical

Abs

orpt

ion.

CC

= Co

mbi

ned

Cycl

e; C

FB =

Circ

ulat

ing

Flui

dize

d Be

d; I

GCC

= I

nteg

rate

d G

asifi

catio

n Co

mbi

n ed

Cycl

e; O

xF =

oxy

fuel

ling;

SO

FC =

Sol

id O

xide

Fue

l Cel

l; U

SC =

Ultr

a Su

perc

ritic

al.

Sour

ce: I

EA G

HG

, 200

0; D

avid

and

Her

zog,

200

1; D

ijkst

ra a

nd Ja

nsen

, 200

3; F

reun

d an

d D

aviso

n, 2

002;

SFA

, 200

2; H

erzo

g, 2

003;

Bra

ndvo

ll an

d Bo

lland

, 200

2; N

saka

la e

t al.,

200

3; B

olla

nd a

nd U

ndru

m, 2

003.

12. T

he IG

CC d

ata

for 2

010

refe

rs to

a E

urop

ean

high

ly in

tegr

ated

pla

nt b

ased

on

a Sh

ell g

asifi

er, w

hile

the

2020

dat

a re

fers

to a

less

inte

grat

ed U

S de

sign

bas

ed o

n an

E-g

as g

asifi

er. T

he e

ffic

ienc

y re

mai

ns a

tth

e sa

me

leve

l bec

ause

new

gas

tur

bine

s w

ill b

ecom

e av

aila

ble

in t

he 2

010-

2020

per

iod

(the

so-

calle

d ‘H

-clas

s’) a

nd re

sult

in a

n in

crea

se in

eff

icie

ncy.

The

gasi

fier s

ubst

itutio

n re

duce

s ca

ptur

e ef

ficie

ncy

loss

esan

d re

duce

s in

vest

men

t co

st p

enal

ties.

037-100 chapter 3 18/11/2004 07:20


their successful development is far from certain. The development of conceptual designs to full-scalepower plants is generally a slow process that will take decades. CCS could be applied in the short term,but the cost and efficiency penalties would be higher than the ones listed in Table 3.4.

The optimal CO2 capture system for gas-fired power plants is not yet clear. New solvents are beingdeveloped that reduce the energy needs for chemical absorption technology. Oxygen-fuelled systemswith CO2 recycle are also being examined. Finally, steam reforming of natural gas with fuel gas CO2

capture, in combination with new hydrogen gas turbines, is being investigated. Overall, it seems likelythat novel approaches, such as re-thinking the power generation process, are needed if substantialreductions in the cost of capture are to be achieved.

Retrofitting CO2 capture technology onto existing power plants

All the designs that have been discussed so far represent greenfield investments. Some studies suggestit might be possible to retrofit power plants with CO2 capture at a later stage.

In a case study of a new gas-fired power plant at Karstø in Norway, two capture systems were compared.The first was an integrated system, where steam was extracted from the power plant, and the secondwith its own steam supply. The integrated system resulted in an efficiency loss of 11 percentage points(from 58% to 47%). The stand-alone system resulted in an efficiency loss of 14.3 percentage points(from 58% to 43.7%). The wider applicability of this option would depend on local gas prices. However,power plant investment costs would be virtually the same at 675 Euros/kW. Given these figures,retrofitting high efficiency power plants may be a feasible option in the future, if gas prices aresufficiently low (Elvestad, 2003).

For IGCCs, it might be possible to reserve space for future expansion with CO2 capture equipment (SFAPacific, 2002). The initial design would accommodate space for a shift reactor, Selexol units, a largerAir Separation Unit, expanded coal handling facilities and larger vessels. Also CO2-capture would involvechanges in the gas turbine, as the gas composition would change. A case study suggests that an initialdesign that considers later retrofit would reduce capture investment cost from 438 USD/kW to305 USD/kW. However, initial investment cost would be 59 USD/kW higher (Rutkowski and Schoff,2003) meaning the net investment cost reduction for IGCC and CCS (in comparison with CCS for notcapture ready IGCC) would be around 17%.

Pulverised coal-fired plants could also be retrofitted, with oxyfuelling seeming the best option (Singhet al., 2003). Total primary energy use for an Air Separation Unit (ASU), low temperature flash (LTF)for purification of CO2 from 95% to 98%, and CO2 separation and pressurization to 150 bar amountto 3.1 GJ natural gas per tonne of CO2, assuming that the electricity needed is produced in a gas-firedcombined cycle. The electricity use for CO2 capture (air separation, CO2 purification and CO2 pressurization)amounts to 35% of the electricity produced in a plant without CO2 capture.

Assuming 40% electric efficiency for the original power plant, additionally 0.72 GJ gas is needed perGJ of electricity produced. The percentage of CO2 avoided is 74%. Capital cost amounts to 120 USD/tCO2 captured (for a 400 MWel coal-fired power plant where 2.7 Mt CO2 per year is captured). Half ofthe capital cost is accounted for by the ASU (Figure 3.1). Assuming an annuity of 15% of the investmentcost, CO2 capture cost amounts to 27 USD/t CO2 captured, or 33 USD/t CO2 avoided.

Lower costs can be achieved for greenfield oxyfuelling plants. One reason is that the process can bedesigned so the CO2 recycle flow can be reduced significantly. Another is that better process integrationcan reduce electricity losses by 6 percentage points (Jordal et al., 2004).

037-100 chapter 3 18/11/2004 07:20

The observations regarding retrofitting are important for a power sector strategy because there seemslittle incentive for CO2 capture in the short term, while there is a need for significant new electricityproduction capacity. Also, coal-fired power plants have a very long lifespan, and existing plants mayneed to be retrofitted if emissions reduction becomes a priority. The relevance of this strategy dependson the age profile of the capital equipment stock.

Efficiency first: clean coal technologiesThe net electric efficiency of individual operational coal-fired power plants ranges from 25% to 48%,and the regional average gross electric efficiency from 27% to 40% (Table 3.5). This wide range canbe attributed to varying steam conditions, coal quality, cooling water temperature and the installationof emission mitigation equipment. A low efficiency power plant can make economic sense when fuelprices are low (as they are in many parts of the world), or when high-efficiency technology would implyimports of costly equipment (the case in many developing countries). CO2 capture from plants withlow electric efficiency makes no sense.

The higher the electric efficiency, the lower the emission mitigation cost will be, and the lower the costincrease per kWh of electricity (see box below). Therefore, investing in high efficiency power plantsis a first step in a CCS strategy. All strategies for increased power plant efficiency are aimed at highertemperature conditions (Figure 3.6). However, higher temperature may also mean more corrosion andhigher steam pressure. These factors constitute materials design problems.

Coal-fired steam cycles can be classified according to their steam conditions: namely, subcritical,supercritical and ultra-supercritical13. Supercritical coal-fired power plants can be considered an established


Figure 3.1

Investment capital cost shares for oxyfuel retrofit of coal-fired power plants

Key point: The air separation unit accountsfor half of the investment costs associated with oxyfuelling

Boiler upgrades1%

Gas combined cycle 24%

ASU48%

Overheads11%

LTF16%

ASU = Air Separation Unit; LTF = low temperature flash for purification of CO2.

Source: Singh et al., 2003.

13. The critical point for water is reached at 401°C and 221 Bar. At higher pressures and higher temperature, there is no longer a discerniblephase transition from liquid to gas.

037-100 chapter 3 18/11/2004 07:20


Figure 3.2

European coal-fired power plant building activity (1920-2000)

Key point: Most standing coal capacity in Europe and North America is over 25 years’ old

GW

/yr

16

14

10

12

8

6

4

2

0

1920

1930

1940

1950

1960

1970

1980

1990

2000

Source: UDI, 2003.

Figure 3.3

North American coal-fired power plant building activity (1920-2000)

GW

/yr

20

18

12

16

14

10

8

6

4

2

0

1920

1930

1940

1950

1960

1970

1980

1990

2000

Source: UDI, 2003.


Figure 3.4

Japanese coal-fired power plant building activity (1920-2000)

Key point: Most standing coal capacity in Japan and China is less than 20 years old

GW

/yr

4.5

3.5

2.5

3

2

1.5

1

0.5

4

0

1920

1930

1940

1950

1960

1970

1980

1990

2000

Source: UDI, 2003.

Figure 3.5

Chinese coal-fired power plant building activity (1920-2000)

GW

/yr

18

16

14

10

12

8

6

4

2

0

1920

1930

1940

1950

1960

1970

1980

1990

2000

Source: UDI, 2003.

037-100 chapter 3 18/11/2004 07:20


Retrofitting ageing, existing power plant stock

The discusison on retrofit can be split into existing plants and new plants that will bebuilt in the coming decades without CCS. The existing power plant stock age differs aroundthe world, depending on the historical electricity demand and supply mix. It is unlikelythat recently-constructed power plants will be closed down within the next couple of decades.This is typically the category where retrofit may be considered. For this reason, the ageprofile of coal-fired power plants has been analysed in more detail here. Figures 3.2 to 3.5show the age profile for four regions: Europe, USA and Canada, Japan and China.

North American stock shows a clear peak around 1970. Given a lifespan of 40-50 years,this suggests that many plants must be replaced around 2010-2020. This is the timeframewhen CCS may become available. Therefore, retrofitting seems less relevant for NorthAmerica. In the European case, only about a third of the stock is under 15 years in age.In 2020, these plants will have a lifespan of almost 35 years. Retrofit may be consideredfor these plants. In the case of Japan and China, the bulk of coal-fired power plant stockis under 15 years of age, meaning that retrofit may be considered. However, large-scaleintroduction of CCS in China is uncertain in the short and medium terms. This leaves Japanas a prime candidate for retrofit, along with certain parts of Europe.

The lifespan of coal-fired power plants is a source of uncertainty. One important factorthat determines power plant lifespan is the outage rate. Unit forced outage rates for coalfired plants are generally low at an age of 10-20 years at about 5%, and increaseexponentially to 20% at an age of 40 years. As a consequence, a trade-off exists betweenlow efficiency and outage of ageing plants, versus investments in new plants (Armor,1996). Generally speaking, the lifespan of US plants is 10-15 years longer than that ofEuropean plants, which may be attributed to lower coal prices and more liberalized marketsresulting in a reluctance to invest in new more efficient plants.

In the USA, repowering projects for existing coal plants have significantly extended plantlifetimes and, in certain cases, resulted in substantial efficiency improvements. The changesare often so substantial that the projects are similar to the entire plant being replaced. In thestatistics, however, such a facility may show up as a power plant with a very long lifespan.

Monitoring of power plant operations has improved, which results in a much better control ofprocess conditions and longer power plant life. However steam temperature and pressurehave increased in order to improve power plant efficiencies. Chemically-reducing conditionsfor NOx-control reasons and changing ash chemistry due to co-firing of other fuels, reduce apower plant’s lifespan (Fleming and Foster, 2001). Electricity market liberalization has resultedin much more start-and-stop cycles than were considered in the original plant design. The neteffect of these changes is a considerably reduced boiler life (Paterson and Wilson, 2002).

The lifespan of coal-fired power plants may change in the future, but it is hard to saywhether it will increase or decrease, compared to plants that are closed down today. Thelifespan of gas-fired power plants is much shorter than coal-fired power plants and capitalcosts are low compared to fuel costs. Therefore, retrofitting gas-fired power plants is unlikely

037-100 chapter 3 18/11/2004 07:20

technology. New coal-fired power plants based on this technology can achieve efficiencies of between44% and 45% (LHV). Even higher values of up to 47% and 48% are reported, but these can beattributed to exceptional conditions with low temperature seawater cooling.

Coal-fired power plants in the USA tend to have lower efficiencies than those in Europe or Japan dueto higher flue gas temperatures at the outlet (the result of the sulphur content in the coal), highercooling temperatures and the use of single versus double reheat. The difference can amount to 3 to5 percentage points (Viswanathan, 2003).

The efficiency of steam cycles is determined by the steam conditions, especially the temperature. For anideal Carnot cycle, the efficiency is given by the ratio of the cycle temperature difference, divided by themaximum temperature of the steam, expressed in Kelvin. So, if a steam cycle has a maximum temperatureof 580°C and a cooling water temperature of 20°C, the Carnot efficiency is (580-20)/(580+273) =


to be a strategy of prime importance. In general, technical lifespan is of secondaryimportance, as power plants can be replaced before the end of their lifespan.

Retrofitting has not been considered in the model analysis presented in Chapters 4, 5, 6and 7 because it is likely to be of limited relevance. Early replacements and reduced loadfactors for fossil fuelled plants without CCS have been considered.

Figure 3.6

Power plant efficiencies as a function of the cycle temperature(based on lower heating value, LHV)

Key point: Higher process temperatures are the key to higher electricity production efficiencies

Effic

ien

y (%

)

90

60

70

80

40

50

30

20

10

05000 1000 1500 2000

Temperature (Degrees Celsius)

IGCC

PFBC

Steam cycle + deNOx and FGD

NGCC

Carnot

Note: The Carnot efficiency is the theoretical maximum, assuming a cooling water temperature of 15°C (Eurocoal, 2003). FGD = FlueGas Desulphurization; PFBC = Pressurized Fluid Bed Combustion; NGCC = Natural Gas Combined Cycle.

037-100 chapter 3 18/11/2004 07:20

66%. In practice, the efficiency is lower (up to 45% efficiency) because of significant losses for flue gascleaning, pumping and other factors. If the steam temperature can be raised by 100°C to ultra-supercriticalsteam cycle (USCSC) conditions, the theoretical efficiency gain is 3 percentage points (48% to 50%).

In turn, the steam’s maximum temperature is limited by materials that can withstand the steamconditions. Current steel alloys have reached their limits at a maximum steam cycle temperatureof about 600°C. Research in the 1990s that focused on ferritic (up to 650°C) and austenitic steels(up to 700°C, so-called P91 and P92 alloys) for higher temperatures has not yielded satisfactoryresults (Fleming, 2002). In recent years, research has focused on nickel alloys. These alloys havethe necessary properties to withstand temperatures of between 700-750°C.

The problem with nickel alloys is their price, which is 10 times that of ferritic and austenitic steels, and100 times more than carbon-manganese (C-Mn) steels (Fleming, 2002). Moreover, the use of significanttonnages of nickel could increase the price of nickel. This has hindered the widespread use of thesealloys so far. New power plant designs are being proposed in which nickel alloys are only used forcritical parts of the power plant. Novel plant designs such as two pass, inverse twin tower and horizontalboiler concepts can reduce investment costs, and have been developed in the framework of the EU700°C power plant project (AD700, not dated; Scott 2001). Their introduction would limit the use ofnickel alloys and enable the construction of power plants with a maximum steam temperature of 700°C.The net electric efficiency of such USCSC plants could exceed 50%. However, the reliability of thesepower plants needs to be proven in practice, a process that will take decades.

Currently, it is not possible to say with a high degree of certainty what the characteristics of futureUSCSC power plants will be. It is, however, possible to define development targets that should bemet by a cost-effective power plant design. Viswanathan (2003) concludes that USCSC total plantinvestment cost could be 12-15% higher than the cost of a subcritical steam cycle, and still be


Table 3.5

Average regional efficiencies for centralized, coal- and gas-fired power plants(2000)

Hard coal-fired (%) Natural gas-fired (%)

Africa 35 36

Australia/New Zealand 38 48

China 35 39

Central and South America 35 39

Eastern Europe 27 44

India 28 44

Japan 38 46

Middle East 40 35

Mexico - 36

Other Asia 33 40

South Korea 36 50

USA/Canada 36 38

Western Europe 39 47

Note: Gross efficiency, excluding own electricity consumption. Based on Low Heating Value, LHV

Source: Based on IEA energy statistics.

037-100 chapter 3 18/11/2004 07:20


Why CO2 capture only makes sense for high efficiencypower plants

It is often argued that CO2 capture should be applied to low-efficiency power plants indeveloping countries as they emit the highest quantity of CO2 per kWh. This reasoning isflawed. Firstly, new types of power plants such as IGCC allow the use of much more efficientlow-cost CCS technology. Secondly, capture equipment costs are subject to economies ofscale. Power plants in industrialized countries are often a factor of two to five times largerthan in developing countries.

The loss of electric efficiency, in relative terms, is much higher for low efficiency powerplants. This can be illustrated by comparing the physical absorption systems of two powerplants: a 35% efficient coal-fired power plant (case A) and a modern 50% efficientpower plant (case B). In case A, 0.96 kg CO2 must be captured per kWh. In case B, 0.67kg CO2 must be captured per kWh.

Assuming that a chemical absorption system is added, steam requirements in case Aamount to 0.0028 GJ/kWh, and to 0.0019 GJ/kWh in case B. Assuming a steamgeneration efficiency of 85%, this represents an increased fuel use of 45% and 32%,respectively. Also, electricity is needed for CO2 pressurization (0.34 GJel/t CO2). Theelectricity output declines by 13% and 8.4% respectively14. Therefore, the fuel use per kWhincreases by 67% and 44% respectively15. This additional fuel use results in additionalCO2 which must also be captured.

Accounting for this increase implies additional fuel use per kWh of 77% and 44%, forcase A and B respectively16. At a coal price of 1.5 USD/GJ, the additional fuel cost amountsto 1.2 US cents/kWh in case A and 0.6 US cents/kWh in case B. Per tonne of CO2 ,additional fuel costs amount to 7 USD/t CO2 in case A and 5.4 USD/t CO2 in case B.17

The capital cost is proportional to the amount of CO2 to be captured, so the capital costper tonne of CO2 will be similar for both plants. In terms of cost per tonne of CO2 andcost per kWh, case B is superior to case A. Clearly, efficiency is the first key step. In conclusion,this shows that retrofitting CCS onto low-efficiency coal-fired power plants in developingcountries is not a viable strategy.

cost-effective. As the balance-of-plant cost is 13-16% lower (because of reduced coal handling andreduced flue gas handling), the boiler and steam turbine cost can be up to 40-50% higher.

Viswanathan concludes that the cost of an USC boiler would be 28% higher than that of acomparable subcritical boiler, but notes that unknowns in fabrication and erection costs with thenew materials could change the results somewhat. Given loss of efficiencies for CO2 capture, theefficiency of USCSC with CO2 capture could reach 42%. This analysis suggests that IGCC development

14. 0.131 = 0.96*1.45*0.34/0.0036*0.001

15. 1.67 = 1.45/(1-0.131)

16. 1.77 = 1+0.67*1.67/1.45

17. 7 = 0.012/0.00096/1.77

037-100 chapter 3 18/11/2004 07:20

might not be crucial for a CCS strategy and that CCS could be applied successfully within the existingsteam cycle technology paradigm.

CO2 Capture in the Manufacturing Industry

Process industriesCO2 capture could be applied in a number of production processes in the manufacturing industry.Industrial sources of (potentially) relatively pure CO2 are of limited importance on a globalscale (<200 Mt CO2 per year in total) and include the production of ammonia, ethylene oxide,existing hydrogen production and production of direct reduced iron (DRI).

Given the limited CO2 capture costs, these processes constitute prime candidates for the introductionof CCS. However, a number of important industrial processes – such as blast furnaces, cementkilns, steam crackers – are characterised by lower CO2 concentrations18 but large CO2 quantities.Because of the low CO2 concentrations, they would require either costly and energy-intensive CO2

chemical absorption processes, or process re-design to increase CO2 concentrations, such as thosebased on the use of oxygen in combination with post-combustion CO2 removal or hydrogen productionin combination with pre-combustion CO2 removal.

Ammonia production

Nitrogen fertilizers are produced from ammonia which is produced from hydrogen. In turn, thehydrogen is produced from natural gas, heavy oil or coal. In older ammonia production plants, CO2

is separated from the hydrogen before the ammonia production step. In newer plants, hydrogenrather than CO2 is separated from the syngas. The residual gas containing CO2, CO, unconvertedmethane etc. is used as a fuel in the reformer furnace, in which case there is no pure CO2 stream.If there was a need to produce pure CO2, it would imply a switch back to the old plant design.

A significant share of the CO2 separated is used for the production of urea (CH4N2O), a populartype of nitrogen fertilizer. Given its chemical formula, 0.88 tonnes of CO2 are needed for each tonneof urea produced. Global ammonia production amounted to 111 Mt in 2000 and urea productionto 46 Mt (UN, 2003). Energy use for ammonia production amounts to 25-40 GJ/t, resulting in aCO2 emission of roughly 1.5 tonnes per tonne of ammonia. If the CO2 needs for urea productionare accounted for, about 150 Mt CO2 could be recovered for underground storage based on year2000 production levels.19 This quantity is relatively small, but it has the advantage that no newcapture process would be needed, simply pressurization. As a consequence, this CO2 is availableat low cost.

Iron and steel production

Substantial amounts of CO2 are captured in the iron and steel industry in the production of DirectReduced Iron (DRI). The bulk of this CO2 is released into the atmosphere. Global DRI productionamounted to 38 Mt in 2001, resulting in CO2 emissions of approximately 20-30 Mt CO2. The productionof DRI is mainly concentrated in countries with cheap stranded gas, including the Middle East.


18. But in many cases still higher than for power plants.

19. A fraction of the captured CO2 is also traded as food grade CO2. However, the quantities are minor.

037-100 chapter 3 18/11/2004 07:20

Iron production in blast furnaces requires approximately 500-550 kg of coke and coal per tonneof product. Total global iron production is about 540 Mt, providing a source of around 1,000 Mtof CO2. Iron production is forecast to decline to around 350-400 Mt in 2030 as secondary steelproduction grows (Gielen and Moriguchi, 2003). The CO2 emission from blast furnaces amountsto 1-1.5 t/t iron. This CO2 can be removed by re-designing the blast furnace for oxygen use andsubsequently removing CO2 using physical absorbents. So far, this strategy has received only limitedattention. Preliminary estimates suggest capture costs in the range of 10-20 USD/t CO2, similarto the capture costs for IGCC (Gielen, 2003).

If CO2 capture was applied to iron and steel production, its potential would be in the order of0.5-1.5 Gt per year. The iron and steel industry is currently studying the best way of reducingemissions. A European project has started, known as ULCOS (Ultra Low CO2 Steelmaking), whichincludes new engineering studies of CO2 capture and sequestration strategies for iron productionprocesses. This project is the European part of the globally-oriented CO2 breakthrough project ofthe International Iron and Steel Institute. Introducing CO2 capture in iron and steel productionmay be hampered by international competitiveness issues, depending on the policy approach chosen(Gielen and Moriguchi, 2003).

Cement

Worldwide, cement kilns emit about 1.3 Gt CO2 per year, equal to 0.6-1.0 t CO2 per tonne of Portlandcement, depending on fuel and energy efficiency. Cement production is increasing, resulting inrising CO2 emissions from this source category. Cement kiln CO2 off-gas concentrations are higherthan for conventional furnaces in other sectors because more than half of the CO2 in the off-gas(0.5 t CO2/t Portland cement) comes from a chemical reaction essential for cement production(so-called calcination): CaCO3 → CaO + CO2

This inorganic portion means that the CO2 concentration in the flue gases is about twice that incoal-fired power plants. Therefore, physical absorption systems (Selexol or other absorbents) couldbe used. Energy-related CO2 emissions depend on the energy efficiency of the kiln (which may rangefrom 3-8 GJ/t cement clinker) and on fuel type (more than half of the fuel may be waste wood,waste tyres etc., which is often not properly accounted for in energy statistics).

So far, no radically new designs have been proposed for cement kilns. The capture technologycould be similar to that of an IGCC or a pulverised coal fired power plant with CO2 capture fromthe flue gas. It might be possible to use oxygen instead of air in cement kilns. However, this wouldimply a process re-design in order to avoid excessive equipment wear. The effects on the processchemistry also need to be assessed. Preliminary data suggests about 0.9 GJel/t CO2 for an oxyfuelprocess with CO2 recycling and 90% recovery efficiency (Hendriks et al., 1999). However, theseare very preliminary estimates and key data, such as the impact of a CO2 atmosphere on thecalcination process, are not known. As a preliminary estimate, the oxyfuel data for coal-fired powerplants is used in the model. Currently, there is no major ongoing research in this area. Instead thecement industry is focusing on energy efficiency, use of waste fuels, and changing resources.

Investment costs have been set at 200 USD/t CO2 annual capture capacity. This should be consideredas a working assumption.

Petrochemicals

The production of ethylene oxide from ethylene inherently produces pure CO2 as a by-product of achemical reaction. Global ethylene oxide production amounts to a few Megatonnes per year, meaning


037-100 chapter 3 18/11/2004 07:20

this is a CO2 source category of secondary importance. Other petrochemical CO2 emission sourcesare steam boilers, furnaces and CHP plants. Capture potential from CHP plants is similar to thatof other power plants, so the discussion of options will not be repeated here. One of the large-scaleprocesses is steam cracking of naphtha and other oil products to yield ethylene and other basicbuilding blocks for the petrochemical industry. In this process, a mixture of residual gas from thecracking process and natural gas is used to heat the furnace of the steam cracker. Residual gas isa mixture of hydrogen and methane, so it is a gas with low CO2 emissions per unit of energy.Chemical absorption is, therefore, the only feasible option.

Paper mills and ethanol plants

Both in chemical pulp production and in ethanol production from ligno-cellulose crops or sugarcane, only the sugar/cellulose and hemicellulose fraction of the plant is used. The remaininglignin fraction (called ‘black liquor’ in pulp processing) can be used for energy recovery. Strictlyspeaking, ethanol production is part of the fuels supply. However, as the CO2 capture process isvery similar to that for black liquor IGCCs with CO2 recovery, it will be described in this section.

Energy recovery from black liquor is an established technology in the pulping industry. Such plantshave a scale of 50-200 MW electric capacity. Currently, Tomlinson boilers are used for energy recoveryand chemicals recovery from black liquor. IGCC technology, which can improve efficiency, is nowbeing tested on a pilot plant scale as an alternative to these boilers. Black liquor IGCC technologyis similar to coal-fired IGCC technology. Such plants could be equipped with CO2 capture. The electricefficiency of a black liquor IGCC is 28%; with CO2 capture it declines to 25%. The steam efficiencyremains at 44% in both cases. Capital cost increases by 320 USD/kW electric when CO2 captureis installed (Möllersten et al., 2003). The technology could also be applied to residues from ethanolproduction, provided future ethanol plants can reach sufficient economies of scale.

It is assumed that 40% of the biomass feedstock for ethanol production ends up as residue. Theresidue can be used for energy recovery, similar to black liquor in the paper and pulp industry. CO2

capture from the gasified residue amounts to 413 kg/GJ electricity, or 104 kg/GJ ethanol produced.Including electricity and heat by-products (18 kg CO2/GJ ethanol) and gasoline replacement(73 kg/GJ ethanol + 7 kg/GJ upstream), the emission reduction percentage compared toconventional transportation fuels amounts to 253%. This means that a 50/50 mixture of ethanolproduced with CCS and gasoline would represent a CO2-neutral fuel.

Furnaces and CHP

Apart from dedicated processes, general boilers and furnaces can be equipped with CO2 capture.CHP systems that represent an energy-efficient alternative to stand-alone boilers and furnaces canbe equipped with CO2 capture as well.

CO2 concentrations in the flue gases of gas-fired boilers are about 7% compared to around 14%in a coal-fired boiler (Thambimuthu, 2003). Because of these low concentrations, chemical absorptionis the only feasible capture strategy. However, as with power plants, oxyfueling may be applied toincrease CO2 concentrations. Pre-combustion reforming, followed by CO2 capture and hydrogencombustion, could also be applied. Such strategies will be limited to large plants (10 MW+). Forsmaller plants, a link to a hydrogen supply system seems much more likely. CCS could be appliedin the centralized hydrogen production process.


037-100 chapter 3 18/11/2004 07:20

Part of the industrial process heat demand in the 400-800°C range can be provided by gas turbines.This option could be particularly attractive in the chemical industry (steam crackers) and in therefining sector (furnaces). Such processes could be equipped with CO2 capture (Table 3.6). Hotexhaust gas from a gas turbine is directed to a furnace of an oil refinery at a temperature of560°C. The gases are further heated to 700-800°C. After heating the oil, the cooled exhaustgases leave the plant at a temperature of about 150°C. The CO2 is recovered from this streamusing a chemical absorption technology. Electricity generated by the gas turbine is sold or usedelsewhere.

The reference case outlined in Table 3.6 consists of a furnace with the same heating capacity andthe power production is balanced by power from the grid (assumed to be a gas-fired combinedcycle with 54% electric efficiency). Total fuel requirements for the CHP unit with CO2 capture areabout 2.57 times as high as for a boiler. However, the unit produces 0.78 GJ electricity/GJ heatas by-product, and the CO2 is captured. Given that a chemical absorption technique is used, it seemsthat the gas turbine uses air. The CO2 removal efficiency is 90%.

Hendriks et al. (2001) indicate a CO2 avoidance cost of 16 USD/t CO2 for such a system. Theseavoidance costs are considerably lower than the avoidance cost for power plants. These costs areexpressed compared to the reference furnace without CHP. Comparing the CHP unit with and withoutCO2 removal would provide a better, albeit higher, estimate of the actual cost.

CO2 Capture in Fuels Supply

The extraction of oil, gas and coal results in almost 400 Mt of CO2 emissions (Table 3.2). The fueltransformation sector is an even more important emissions source. Petroleum refineries and LNGproduction account together for 700 Mt of CO2 emissions per year (Table 3.2). In future, theseemissions are bound to increase significantly. On the fuels supply side, LNG production will increasesignificantly, as larger quantities of natural gas must be transported over longer distances wherepipelines do not constitute a viable alternative.


Table 3.6

Characteristics of furnace/CHP unit with CO2 capture

Reference plant CHP with CO2 separation

Capacity furnace (MWth) 450 450

Load (hrs/yr) 8,200 8,200

Investment cost (million USD) - 214

Fuel and O&M (million USD/yr) 35 101

Efficiency CHP - 35% (electric) / 45% (heat)

CO2 production (Mt/yr) 1.57 1.61

CO2 recovered (Mt/yr) - 1.4

Electricity costs (USD/kWel) - 0.025

Costs (USD/t CO2 avoided) - 16

Note: Gross efficiency, excluding own electricity consumption. Based on Low Heating Value, LHV

Source: Hendriks et al., 2001

037-100 chapter 3 18/11/2004 07:20

Currently, emissions from oil product use exceed the emissions from oil production and processingto a considerable extent. However, this may change in the future. Heavier crude oil types that requiremore upgrading are likely to gain market share as the quality of the remaining oil reserves declines.Synfuel production (e.g., through Fischer-Tropsch synthesis) is considerably more energy intensivethan conventional refining. The use of hydrogen as a transportation fuel would result in the possibilityof zero vehicle tailpipe emissions, and a significant potential to capture CO2 from hydrogen production.Synfuels are projected to gain an increasing market share. Synfuels such as hydrogen, methanol,dimethylether, and synthetic gasoline and diesel can be produced from natural gas, coal or biomass.CO2 capture could be applied to these production processes.

This section will discuss the following four categories of CO2 capture from fuel supply:

● CO2 capture in natural gas processing;

● Refinery CO2 capture;

● Hydrogen production processes;

● Gasification and Fischer-Tropsch production of synfuels.

Natural gas processingThe CO2 content of natural gas varies from virtually zero in Siberian gas, to 1.5% in certain NorthSea gas fields and up to 70% in fields such as Natuna in Indonesia. The latter value is an extreme;an average CO2 content is 1-2%. The quantity of CO2 that is released when the gas is combustedis two orders of magnitude larger than the CO2 from gas processing. This limits the worldwidepotential for CO2 capture in natural gas processing to less than 100 Mt CO2 capture per year.

CCS for natural gas processing projects is receiving much attention as a low-cost CCS opportunity.CO2 must be removed anyway before the gas can be sold, and storage sites are often nearby. Theadditional costs for compression, transportation and storage are limited. Moreover, CO2 storagewells are similar to gas production wells, so the necessary equipment and expertise are availableon site. Most existing and planned CCS projects are gas production projects. These include theSleipner and Snohvit projects in Norway, the In Salah project in Algeria, the Gorgon project inAustralia and the Natuna project in Indonesia.

Oil refineriesOil refineries convert crude oil into oil products. They do so through a wide range of process operations.The most important of these are distillation, reforming, hydrogenation and cracking. Distillationprocesses require low temperature heat; hydrogenation requires hydrogen, and cracking producessignificant amounts of heat and CO2 from heavy oil residues. Refineries also consume considerableamounts of electricity. A subdivision of CO2 emission sources of two types of refineries is shown inFigure 3.7.

Reformers, fluid catalytic crackers (FCCs) and possibly vacuum distillation units could be equippedwith high-temperature CHP units with CO2 capture. Together they represent 30-40% of the refineryenergy consumption. On average, 5-10% of the crude throughput of refineries is used for the refiningprocess. Modern refineries have higher emissions because they can use heavier crudes and producemore light products, especially gasoline and diesel.

Refinery heaters can be equipped with post-combustion CO2 capture technology. A study for a UKrefinery and petrochemical complex suggests that collecting 2 Mt of CO2 per year would require10 MW for blowers to push the flue gas through the network, and 10 MW for the pressure drop


037-100 chapter 3 18/11/2004 07:20


Figure 3.7

CO2 emissions from oil refining

Key point: Process heaters account for half of the CO2 emissions from oil refining

Process heaters44%

Power13%

Hydrogenplant 20%

Utilities23%

FCC refinery

Process heaters55%

Power14%

Hydrogenplant 14%

Utilities17%

Hydrocracking refinery

Sources: American Petroleum Institute, 2002; Clarke, 2003.

Figure 3.8

Investment cost structure for a refinery complex with CO2 capture

Key point: CO2 separation and compression is responsiblefor less than half of the capture investment costs for oil refining

NOx and SO2 removal16%

CO2 separation35%

CO2 drying and compression10%

Utility and offsite systems 31%

Gas gathering systems8%

Source: Simmonds et al., 2003.

037-100 chapter 3 18/11/2004 07:20

imposed by the packed column absorbers (Simmonds et al., 2003). This equals 0.39 GJ/t CO2. Pre-treatment is needed to reduce NOx and SO2 concentrations. The system needs 396 MW of naturalgas, which amounts to 6.2 GJ natural gas per tonne of CO2 captured. This includes the energy needsfor the blowers and the steam for the regeneration of the absorbents. This is a fairly high energyconsumption, compared to CO2 capture energy needs for power plants. There may be room forfurther improvements in the design. The investment costs amount to 238 USD/t CO2 with theoperational cost largely determined by natural gas costs. A breakdown of the investment costs isshown in Figure 3.8. Note that this is a conventional refinery. There may be potential to reducecapture cost through synergies, such as by using refinery waste heat for CO2 capture.

Another study focused on oxyfuelling for a refinery power station boiler, using heavy oil and gas(Wilkinson et al., 2001). Electricity needs for the air separation unit and CO2 separation amountto 1.5 GJ electricity per tonne of CO2. Investment costs would amount to 50 USD/t CO2.

The product mix of refineries is changing towards more light products with a higher H/C ratio, asdemand growth is concentrated in transportation markets. The refineries can respond to the hydrogendeficiency by adding hydrogen (a process called hydrocracking) or by removing carbon (a processcalled coking). This trend is apparent if the regional refinery structures are compared. The higherthe transportation fuel demand as a share of total fuel demand, the higher the coking andhydrocracking capacity (Table 3.7).

Refinery coking capacity is much higher in the USA than in other world regions, while hydrocrackingis concentrated in other OECD member countries and the Middle East. Global hydrogen use forrefineries is already substantial, about 2 EJ in 2000 (0.5% of global primary energy use). Given acrude oil consumption of about 150 EJ (see Table 3.1), this equals on average 1 kg CO2/GJ crudeoil processed. This emission level is bound to rise significantly. For example, in the case of flexicoking,the process emission amounts to more than 20 kg CO2/GJ of fuel processed.


Table 3.7

Regional refinery structure (2000)

Crude Crude Coking Catalytic Gasoline and Comment(Mbbl/d) (Index) (Index) hydrocracking diesel in refinery

(Index) product mix (%)

Australia 0.95 100 0 3 78

Canada 1.91 100 2 14 72

Eastern Europe 1.88 100 4 4 61

FSU 8.40 100 3 1 48 Heavy crude

Japan 4.96 100 2 3 51

Korea 2.56 100 1 5 34

Middle East 5.99 100 1 10 41 Heavy crude

Mexico 1.53 100 3 1 47 Heavy crude

USA 16.54 100 13 9 71

Western Europe 14.90 100 2 6 63

Developing countries 21.64 100 4 3 44-55

World 81.25 100 5 5

Note: Index crude distillation = 100. FSU = Former Soviet Union.

Source: Oil & Gas Journal Energy Database, 2001.

037-100 chapter 3 18/11/2004 07:20

According to IEA statistics, energy used in refining amounted to 11.9 EJ in 2000 (see Table 3.1).About half of this was for natural gas and refinery gas, and the other half for heavy oil products.Worldwide refinery CO2 emissions amount to 0.75 Gt CO2 per year (Table 3.2).

Gradually crude oil quality is changing towards more heavy product types. Unconventional oilproduction is also growing. Canadian oil sands and Venezuelan Orinoco tar sands constitute almost2% of global oil production. These unconventional crude oil types require special refining operationsto adjust the hydrogen/carbon (H/C) ratio. The reserves in place are of a similar order of magnitudeto the quantities of conventional oil, with 580 billion barrels of recoverable reserves (IEA 2002a,p. 101). Total upgraded crude from both sources is projected to increase to 6.1 million barrels perday in 2030 (IEA 2002a, p. 102).

Canadian oil sands production amounted to 829 kbbl per day in 2002, 1% of the total global oilproduction. Crude bitumen extracted from oil sands is refined to a marketable hydrocarbon product20

through a combination of carbon removal in high temperature coking vessels and by hydrogenaddition in high temperature, high pressure hydrocracking vessels. The remaining fraction is eitherthermally cracked to gaseous products or converted into petcoke. The bulk of the petcoke is burnedfor energy recovery. The upgrading processes yield 0.84 cubic metres of syncrude per cubic metreof crude bitumen (Imperial Oil, 2000). The upgrading energy efficiency is 74% and the net emissionamounts to 22-34 kg CO2/GJ syncrude. A typical plant has a capacity of 250,000 bbl per day,amounting to an emission of 18 Mt CO2 per year.

With the Orinoco tar sands, current plans are to apply deep conversion technology in order toproduce high-value transportation fuels. Delayed coking is the primary conversion technology. Plansare to produce 622,000 barrels per day of syncrude by 2009. Four strategic associations havealready started operating and aim for different levels of tar sand upgrading. Of this only theSincor project, based on delayed coking, will be discussed in more detail. Here, some 212 kbbl perday virgin crude are upgraded to 186 kbbl of a 32°API quality synthetic crude (i.e. crude oil witha density of 0.865 t/m3). The product mix consists of 2% LPG, 13.9% naphtha, 17.5% kerosene,


20. The syncrude gives good yields of kerosene and other middle distillates, so it is not exactly the same product as conventional naturalcrude oil.

21. Excludes electricity use for pumps, etc. With coal, the efficiency to liquid products is 41.1% with the power export amounting to 5%of the coal input.

Table 3.8

CO2 emissions in various refining and synfuel production processes

Efficiency21 (%) CO2 available for storage CO2 available for storage(kg/GJ product) (Mt/yr/plant)

Syncrude oil/tar sands 74 34 18

Flexicoker 84 24 5.4

FT natural gas 57-70 7-25 0.25-2

FT coal 40 160 10-15

FT biomass 40 210 0.2

Methanol/DME from coal 65 110 5-10

Methanol/DME from natural gas 70 8 0.25-0.5

Note: FT = Fischer Tropsch synthesis.

Source: Steynberg and Nel, 2004; IEA data.

037-100 chapter 3 18/11/2004 07:20

28.7% diesel, 22.9% gasoil, 14.9% coke and sulphur. Hydrotreating is being considered as analternative to delayed coking (Paez et al., 2000).

Table 3.8 shows emissions of CO2 per GJ of product and emissions per process unit. The high CO2

emissions in oil sand and tar sand production and processing pose a problem. CO2 capture andsequestration may be applied to the residue treatment, thus reducing CO2 emissions to a considerableextent.

Hydrogen production

Hydrogen is a CO2-free energy carrier. Like electricity, it can be produced from any other primaryenergy carrier, either by direct conversion or by production of electricity and subsequent waterelectrolysis. As a result, it is not subject to the same supply security problems as oil. When hydrogenis produced from carbon-containing energy carriers, CO2 and hydrogen must be separated to producepure hydrogen. An energy system based on hydrogen could be CO2-free. Hydrogen productionfrom fossil fuels with CO2 capture could be the first step strategy toward a hydrogen economy,followed by hydrogen production from other CO2-free primary energy sources in the longer term.

Hydrogen is widely considered to be the transportation fuel of the future. The competitiveness ofhydrogen as a transportation fuel depends critically on the cost of hydrogen vehicles and theefficiency gains compared to conventional vehicle engines (IEA, 2003c). Major cost reductions areneeded. Also, there is a chicken-or-egg problem: no vehicles without fuel supply, and no fuel supplywithout demand.

Apart from being a transportation fuel, hydrogen can be used for decentralized electricity productionand for space heating, if a distribution system is in place. Hydrogen fuel cells may gain a marketshare. Given their high electric efficiency in combination with their small size, they would be suitablefor residential and commercial heating systems. However, the discussion in the box below suggeststhat a future trend towards decentralized power production is by no means a certainty.

Least-cost hydrogen supply options with low CO2 emissions are based on fossil fuels with CO2 capture(Figure 3.9). All production routes that involve electrolysis are considerably more expensive: the two-step approach of electricity production followed by hydrogen production incurs higher capitalcosts, and reduces efficiency. This poses a major hurdle for any hydrogen economy built on renewables,except biomass and concentrated solar heat as they would not involve electrolysis. While technologylearning can reduce the cost of hydrogen production from renewables, its cost will remain prohibitivein all but a few regions with abundant cheap renewable energy, such as Iceland.

One of the more exotic options for CO2 capture is hydrogen production from biomass (Read andLermit 2003; Azar et al. 2004). This strategy reduces atmospheric CO2 concentrations and producesenergy at the same time. The scale of biomass hydrogen production is typically one order of magnitudesmaller than coal-based hydrogen production. Given that investment costs of chemical plants typicallyincrease with scale by a factor of 0.7, the specific investment costs for biomass-based hydrogenproduction are twice those of coal-based hydrogen production per unit of energy.

Hydrogen and electricity can also be co-produced from fossil fuels. This reduces the product gasseparation cost and can increase the plant load factor, while improving the economies of scaleand reducing CO2 capture cost. A demonstration project is planned in the USA, known as FutureGen.Synfuel cogeneration is considered in the ETP model (see box below).


037-100 chapter 3 18/11/2004 07:20

The supply cost of hydrogen depends critically on the supply volume. Distribution and refuellingwill add 7-9 USD to the production costs in Figure 3.9. These costs apply to large scale systems.In a transition period, decentralized production and/or liquid hydrogen distribution may be used.The costs of such supply systems are much higher. Moreover, decentralized production systemscannot be combined with CO2 capture and sequestration.

Note that the costs in Figure 3.9 represent technology costs. In practice, investors may demand amuch higher return on capital for new risky investments in a hydrogen infrastructure than forestablished oil and gas-based energy systems.

Gasification and Fischer-Tropsch production of liquid synfuels Gasification of carbon-containing feedstocks, followed by hydrocarbon synfuel production, has receivedmuch attention in recent decades, given the potential for the production of synthetic transportationfuels to reduce dependency on oil. Coal, natural gas and biomass could be used as feedstocks. Anumber of synfuels have been proposed: methanol, DiMethyl Ether (DME), naphtha/gasoline anddiesel. The energy efficiency of the production processes for these fuels ranges from 40% to 70%.As a result, they emit a large volume of CO2 which could be captured and stored.


Figure 3.9

Hydrogen production cost for a fully developed supply system

Key point: Producing hydrogen from fossil fuels with CCSis less costly than doing so using nuclear or renewable energy

USD

/GJ

H2

50

40

30

20

10

0

Natur

al ga

s + C

CS

Coal +

CCS

Biom

ass

Nuclea

r

HTGR

coge

nera

tion

Onsho

re w

ind

Offsho

re w

ind

Solar

ther

mal

Solar

PV

HTGR = High Temperature Gas-cooled (nuclear) Reactor.

037-100 chapter 3 18/11/2004 07:20

Fischer-Tropsch (FT) production of synfuels is an established technology. Production of gasolineand diesel from coal was developed in Germany during the Second World War and further developedby Sasol in South Africa during the oil boycott in the 1980s and 1990s. Shell has a plant inSarawak (Malaysia) that uses similar technology to convert so-called ‘stranded’ gas into longer chainhydrocarbons. The technology is based on fuel gasification to a mixture of CO and H2, followed bycatalytic chain building. The product mix consists of condensate and predominantly wax. The waxcan be cracked to yield diesel and gasoline. The product mix depends on the process condition andcatalyst choice (Zhou et al., 2003). In the ETP model analysis, a 50/50 yield of diesel and gasolinewas assumed.

Gas to liquids (GTL) is currently the most attractive FT option. Up to 1 million barrels per day areexpected to come on stream before 2010, in locations with stranded gas such as Qatar and Nigeria(Chemical Market Reporter, 2004). All these plants produce primarily diesel. Investment costs arecoming down rapidly, mainly because of economies of scale. For example, Sasol claims investmentcosts of 12 USD/GJ per year for new plants of 60 PJ product per year, with a further cost reductionpotential to 9 USD/GJ per year (Marriott, 2000). In the IEA World Energy Outlook Reference


Cogeneration of electricity and synfuels with CO2 capture

The production of synfuels is based on gasification, followed by purification and synthesis(except for H2, where no synthesis step is needed). The gasification step is also applied inIGCC power plants. Therefore, co-generating electricity and synfuels is a logical step andcan lead to economies of scale and higher capacity factors for the gasifier. The reactoroutput has to be separated into product and unreacted feedstock, which must be purifiedand recycled. This separation is a costly and energy-intensive step. Once-through processeshave consequently received much attention.

In recent years there has been increasing attention for co-production processes of electricityand synfuels such as methanol, Fischer- Tropsch diesel and hydrogen from coal. The reasoningbehind these concepts differs. In Europe, the main attention is focused on the comparativelylow average load factor of IGCC power plants. Co-production would allow a high averageload factor, which would reduce capital cost per unit of product (Lange et al., 2001). Astudy by Sasol points out that the co-production of liquids and electricity raises the energyconversion efficiency from 40 to 50%, compared to the same plant without electricitycogeneration (Steynberg and Nel, 2004).

US studies start from IGCC and focus on the supply security benefits and air pollutantbenefits of co-production of synthetic fuels (Gray and Tomlinson, 2001). A coal-basedpower plant for co-production of electricity and hydrogen, in line with the US’s FutureGenproject, has been considered in the model. For gas, such a complex design would makelittle sense, given the comparatively low capital cost of gas-fired power plants and FT-synthesis from gas. Generally, the assessment of co-production is complicated by the factthat both products compete in volatile markets where future prices are hard to predict.Static analysis predicts that synfuel production costs may be reduced by 10% if a co-production strategy is applied (Yamashita and Barreto, 2003). This suggests there maybe benefits to such cogeneration strategies, but they are not crucial.

037-100 chapter 3 18/11/2004 07:20

Scenario, gas-to-liquids is forecast to increase to 2.3 million barrels per day by 2030, some 2% ofworld oil supply (IEA 2002b).

Production of FT transportation fuels from coal with CO2 removal has been described by the CoalUtilization Research Council (2002). Currently, a 40% liquid product yield (in energy terms) canbe attained. The amount of CO2 available for capture is much higher for coal-based processes thanfor gas-based ones (Table 3.8). The energy requirements for CO2 capture are proportional to thequantity of CO2 in the flue gas. Given a gas price of 0.5 USD/GJ, current FT supply costs are 25-30 USD/bbl (Marsh et al., 2003). The capital cost for a coal-based process is about twice that ofa gas-based process. Moreover, the energy efficiency is also lower. The production costs startingfrom coal are twice as high at the same feedstock price. However, cogeneration of fuels and electricitycan reduce these costs (Steynberg and Nel, 2004). Oil price hikes can make coal or gas-based FTtransportation fuel production a viable alternative.

Biomass feedstocks are technically feasible (Ree, 2000). Investment costs for FT biodiesel plantwithout capture are projected to decline from 60 USD/GJ in 2000 to 36 USD/GJ by 2020. Thisis twice the investment costs for coal because of the smaller scale plants. A plant would use 2 GJbiomass and 0.03 GJ electricity per GJ product. At a biomass feedstock price of 4 USD/GJ, thetransportation fuel production cost in 2020 is 15 USD/GJ. This is about three times the currentproduction cost of gasoline and diesel. CO2 capture would add 0.05 GJ electricity use per GJ fuelproduced (including CO2 pressurization). Investment costs would increase by 30% (Marsh et al.,2003). About 120 kg of CO2 is captured per GJ of fuel produced. The net emission reduction,compared to diesel and gasoline from crude oil, amounts to 264%. The emission reduction in excessof 100% is explained by the sum of the replacement of fossil fuels and storage of CO2 from theprocess flue gas. The emission mitigation costs amount to 60 USD/t CO2, but this depends criticallyon the assumed biomass feedstock cost.

DiMethyl Ether (DME) can be used as a fuel for power generation turbines, diesel engines or as anLPG replacement in households. However, its main use is as an aerosol propellant for hairspray.Current global DME production amounts to 0.15 Mt/yr. The only plant that produces DME for fueluse started operation in 2003 in Luzhou, China. This plant has a capacity of 10 kt/yr, and usescoal as a feedstock. A large number of pilot projects are being studied worldwide. Emissions fromthe production of DME from coal amount to 71-75 kg CO2/GJ product. In comparison, emissionsfrom DME production from natural gas amount to 6-16 kg CO2/GJ (Sakhalin Energy, 2004).

Current DME production takes place in two-steps. Methanol is produced from syngas and themethanol is catalytically dehydrated to DME. New production processes are under developmentwhere DME is produced directly from syngas in a single step. Various process designs exist basedon liquid phase conversion or gas phase conversion. Liquid phase conversion is preferred for syngasflows with a high CO content, while gas phase processes are preferred for syngas with a highhydrogen content (such as from natural gas steam reforming; Air Products, 2002).

Various designs have been proposed for methanol/DME co-production and for cogeneration ofDME and electricity. Such designs circumvent the problem of recirculation of products because ofincomplete conversion of feedstock into DME. For example, at 50 bar and 300°C, a CO conversionof more than 50% and a DME selectivity of more than 90% are obtained (Air Products, 2002;Ogawa et al., 2003; Sinor, 2004). In the ETP model analysis, a once-through DME and electricitycogeneration plant with CO2 capture is considered (Larson, 2002). This plant achieves 17% efficiencyin coal-to-electricity conversion, and 33% efficiency in coal-to-DME conversion (5.88 GJ coal input/GJelectricity produced, and 1.94 GJ DME produced/GJ electricity produced).


037-100 chapter 3 18/11/2004 07:20

Technology Learning Effects for CO2 Capture

Technology learning is a term applied to the phenomenon of unit costs of technologies decreasingover time. For a large number of technologies it has been observed that unit production costs declineby a fixed percentage for each doubling of cumulative installed capacity. This observation iswidely used to project future costs of energy technologies (IEA, 2000).

A cost decline can be attributed to various mechanisms. R&D may result in new processes to makethe same product using fewer natural resources. Also, learning-by-doing plays an important role.Producers of equipment find more efficient ways to produce equipment as their experience increases.Standardization helps to reduce unit production cost. Finally, economies of scale play a key role.The larger the process or the equipment, the lower the unit production cost. Engineering literatureusually suggests a 20% cost reduction for a doubling of the unit capacity.

Technology learning is very important for emerging technologies, where the cumulative capacitiesare small and capital cost dominates total process cost. This is the case, for example, for solar PVsystems. The situation for CCS is fundamentally different. First, the energy efficiency loss for CCSand the related additional fuel use represent an important part of the CCS cost. Cost reductionsbased on energy efficiency improvements are usually not covered by technology learning curves,but they should in the case of CCS. There are examples of energy efficiency improvements for existingprocesses, e.g., for chemical absorption of CO2. Second, the process equipment used for most CCStechnologies has been widely applied; the main challenge is the integration of this equipment.However economies of scale may apply. Also, serial production of CO2 compressors may result insignificant equipment cost reductions. The potential for learning-by-doing is probably more limitedthan the potential for learning-by-innovation, but it is not negligible.

Moreover, CCS is not a single technology, as it covers a wide range of technologies. While thetechnology learning potential may be very important for certain CCS relevant technologies suchas fuel cells or separation membranes, these technologies are currently far more expensive thanchemical absorption systems. Applying the learning rate of such CCS emerging technologies toCCS as a whole, starting from the cost for chemical absorption systems, would result in a significantoverestimation of the learning potential.22

Cost reductions for CCS can be split into:● Creating benefits via CO2 use for enhanced fossil fuel production;

● Reducing energy losses for CO2 capture, based on new technology (this reduces fuel cost but itmay also reduce capital cost, especially for power plants);

● Economies of scale;

● Standardization.


22. Riahi, Rubin and Schrattenholzer (2003) have assessed the impact of learning effects for CCS technologies. They assume a progressratio of 87% (according to the authors this is a conservative estimate compared to other emerging technologies, based on learning fordesulphurization technologies). The cumulative capacity is 1 GW for the starting year, with initial costs amounting to 45 USD/t CO2

for capture from coal-fired plants and 30 USD/t CO2 for capture from gas-fired plants (excluding transportation and sequestration).They assume that by the end of the 21st century, 90% of all power plants will be equipped with CO2 capture. According to the learningcurve theory, this would result in a cost reduction by a factor of four. However, the uncertainty surrounding this projection is significant.For example, the 1 GW initial cumulative capacity can be questioned. About 100 Mt ammonia is produced annually and 150–200 MtCO2 is captured in the process. The cumulative ammonia capacity is 300-400 Mt CO2. A similar cumulative capacity of hydrogen productionwith CO2 capture exists in other industries. Total cumulative capacity for ammonia and hydrogen equals 80-110 GW (coal-fired) powerplants. If the learning analysis takes this higher initial capacity into account, the cost reduction potential is significantly reduced. Moreover,it is not clear why energy losses (operational costs) would decline proportionally with investment costs.

037-100 chapter 3 18/11/2004 07:20

In the ETP model analysis, cost reductions are estimated exogenously through specific technologiesand vintage models for technologies with different investment costs for each vintage (Table 3.4).This is considered a sufficiently accurate method, as long-term cost can be projected based onequipment material cost, maximum feasible power plant size and the like. The data in Table 3.4implies a halving of CO2 capture costs between 2010 and 2030, compared to the same powerplants without CO2 capture. The feasibility and future investment cost of speculative technologiesare uncertain.

CO2 transportation

CO2 can be transported via pipelines, by tank wagons and by ship. In practice, because of thehuge volumes involved, only pipelines and ships are cost-effective options. Costs depend on thedistance and volumes involved. Generally, transportation costs are considered to be small comparedto the overall capture costs. Transportation cost estimates range from 1 to 10 USD/t CO2, providedthe pipeline transports more than 1 Mt of CO2 per year and the distance is less than 500 kilometres(Figure 3.10). Per unit of weight the costs for CO2 transportation are much lower than for naturalgas or hydrogen transportation because CO2 is in a liquid or supercritical state, with a 10 to 100times higher density. Therefore, per unit of weight, CO2 transportation is more akin to oil transportationin terms of cost than to natural gas or hydrogen.

An engineering study for the Norwegian Kårstø gas-fired power plant suggests that pipelinetransportation represents about 40% of total CCS project investment costs (Elvestad, 2003). This


Figure 3.10

Cost of overland transportation of CO2 by pipeline

Key point: CO2 transportation costs depend strongly onthe quantities and, to a lesser extent, on the distances involved

USD

/t C

O2

70

60

40

50

30

20

10

01000 200 300 400 500 600 700 800

km

1.75 Mt CO2/yr

1.25 Mt CO2/yr

3.5 Mt CO2/yr

0.5 Mt CO2/yr

0.15 Mt CO2/yr

Source: IEA GHG, 2002a.

037-100 chapter 3 18/11/2004 07:20

is an exceptional case with an over-sized pipeline that has to be laid above ground because of thegeological conditions, but it shows the importance of local conditions for the transportation cost.Cost may be 2-4 times higher than in Figure 3.10 in instances where there are unfavourable conditions,in which case they cannot be neglected. CO2 transportation in itself poses no special safety risks,provided the equipment is appropriately sited and regulations adhered to (Vendrig et al., 2003).

Given potential pipeline siting constraints and transportation distances of hundreds of kilometers,a CO2 transportation ‘backbone’ may be needed to which multiple power plants and a number ofstorage sites can be connected. Such a system would allow transportation over longer distancesat acceptable cost.

Pipeline materials may be corroded by a combination of SO2 and water (that yields sulphuric acid).Sulphur will be converted to H2S in an anaerobic environment (e.g., a CO2 capture process in anIGCC installation) and oxyfuelling will result in SO2. The CO2 purity constraints may necessitatecostly sulphur removal that will add to the cost. H2S can be co-injected without problems, as it hasbeen done on Canada for some years.

Shipping CO2 is an established technology on a kilotonne scale. Shipping may become an importantissue because the prime locations for underground CO2 storage are unlikely to coincide with CO2

source locations. For example, the bulk of the conventional oil reserves are located in the MiddleEast (see next section) and the main gas reserves in the Middle East and Russia. By contrast, themain emission sources are in major population centres of OECD countries. Future emission growthwill be concentrated in developing regions such as eastern China. Therefore, the mismatch of sourcesand sink locations constitute a limitation for underground CO2 storage in depleted oil and gas fields,unless cost-effective inter-regional transportation systems are developed. With regard to enhancedcoal-bed methane recovery (ECBM), coal reserves are more evenly spread across the globe with somereserves close to main population centres.

Liquid CO2 has a density of 1-1.15 tonnes/m3 compared to 0.454 tonnes/m3 for liquefied naturalgas (LNG). Storage tanks for CO2 are made of a less expensive material because transportationtakes place at temperatures of -50 °C, compared to -162 °C for LNG. Current intercontinentalshipping and storage costs for CO2 would be in the range of 25-50 USD/t CO2, based on naturalgas shipping costs.23 Further cost reductions may be achieved. Given these cost levels, it may makesense to ship CO2 for EOR, if the CO2 is provided for free and no CO2 sources exist closer to theEOR site. This means that transportation of CO2 to the Middle East should be considered as along-term option if far-reaching CO2 policies are implemented. Interregional transportation forstorage in depleted oil and gas reservoirs has no obvious advantages over storage in local aquifers.24

The quantities of CO2 involved are large, compared to total world commodity transportation. Globaloil production and shipment amounts to 3.5 Gt per year, global coal production and shipment to3.8 Gt per year, global cement production to 1.6 Gt per year, and global cereals production andshipment to 2.1 Gt per year. In the long run, total CO2 shipment could be of the same order ofmagnitude as shipments of all existing commodities put together. Therefore, the challenge ofputting in place an appropriate transportation system for CO2 should not be underestimated.


23. The largest LPG tanker built to date has a capacity of 22,000 m3 and cost 50 million USD (IEA GHG, 2002). The transportationcost for a 500 km distance would be around 10 USD/t CO2. This would be on par with offshore pipelines, but more than twice the costfor onshore pipelines. However from a distance of 1500 km, shipping seems cheaper. Recent analysis indicates cost of 25 USD/t CO2

for a distance of 6,000 km, if a ship were built now (IEA GHG, 2004b). There would, in addition, be expenditure for a CO2 holding tankat the port, as well as operating expenses. For longer distances, cost per kilometre would decrease. In the ETP analysis it is assumed thatcosts decline to 15 USD/t CO2 for transportation over 5,000 km.

24. This is from a technical point of view. There may be benefits from a social acceptance point of view.

037-100 chapter 3 18/11/2004 07:20

CO2 storage

This section will focus on the following storage options:

● CO2 enhanced oil recovery (EOR);

● CO2 enhanced gas recovery (EGR);

● CO2 enhanced coal-bed methane recovery (ECBM);

● Storage in depleted oil and gas fields;

● Storage in deep saline aquifers;

● Other storage options.

Most studies suggest that injection well costs are of secondary importance compared to the costsof capturing and transporting CO2. This is only correct if the cost for CO2 compression is allocatedto CO2 capture. This is the approach chosen for this study, where the energy efficiency of the captureprocess accounts for pressurization to 100 bar. The injection pressure and transportation distancedetermine the need for CO2 pressurization. The injection pressure is a function of the injection depthand the pressure profile in the ground. The minimum storage depth is about 800 metres (at thisdepth CO2 will be in its supercritical state25), but many gas reservoirs are located 2-4 kilometresbelow ground. For storage at a depth of 800 metres, pressurization to 100 bar may suffice. However,storage in deep depleted gas fields will require surface pressures of 200-300 bar.

The pressurization energy requirement depends on the efficiency of the compressor (between 75%and 85%). Recycling CO2 can also require significant energy. In practice, in the case of EnhancedOil Recovery (EOR), recycling amounts to 16-40% of the CO2 quantity injected (20-67% of theCO2 retained). In the Enhanced Gas Recovery (EGR) case, it amounts to 14% of the CO2 retained.The recycled volume for ECBM may be higher if the coal-bed gas contains significant amounts ofCO2, or if the CO2 by-passes the coal through fractures in the rock. As a working assumption, CO2

recycling for CBM is assumed to amount to 20% of the CO2 retained.


Table 3.9

CO2 pressurization energy requirements for injection as a function of type ofreservoir and depth

100 bar pressure at 800m 200 bar pressure at 1600mof depth (GJel/t CO2) of depth (GJel/t CO2)

Aquifer/depleted oil and gas fields 0.22 0.38

EOR 0.34 0.50

EGR 0.25 0.40

ECBM 0.25 0.40

25. In the supercritical state there is no discernible transition from a gaseous to a liquid state, as pressure increases. The density of CO2

at 100 °C and 200 bar is about 0.5 t/m3, at 500 bar 0.8 t/m3). The CO2 density plateau depends on the local subsurface pressureand temperature gradient, and may range from 0.61-0.72 t/m3 (Rigg, 2001).

037-100 chapter 3 18/11/2004 07:20

Table 3.9 provides an overview of pressurizing requirements. In most studies this electricity use isaccounted for under CO2 capture (as in Table 3.4), because pressurization takes place before theCO2 enters the pipeline.

CO2 injection well costs are small compared to capture and transportation costs. Figure 3.11 showsthe cost of onshore oil wells.26 These costs increase exponentially with depth. A similar cost curveapplies to CO2 injection wells. While injection may take place in an aquifer at 800 metres depth,a depleted gas field may require a 4000 metre-deep well. The cost of the injection well will differby a factor of between 5-10.

In recent years, much attention has been paid to technologies which enhance the production ofoil, gas and coal-bed methane. These CO2 storage options could create additional benefits byenhancing fossil fuel production. The main characteristics of these benefits are listed in Table 3.10.They amount to 1-55 USD/t of CO2 (excluding the costs for the wells and CO2 recycling).

EOR creates the highest benefit, followed by ECBM and EGR. Given CO2 capture costs of 10-35 USD/t (Table 3.10), there is potential to offset part or even total capture costs. In mostcases, the costs for CO2 capture will exceed the benefits of enhanced fossil fuel production.Also, the potential for enhanced fossil fuel production is limited by the location of suitable geologicalformations and the location of CO2 sources.


Figure 3.11

Onshore oil well cost as a function of well depth

Key point: Well costs increase exponentially with well depth

10

00

USD

/wel

l

3000

2000

2500

1500

1000

500

010000 2000 3000 4000 5000

Depth (metres)

Source: Caldwell, 2001

26. Offshore wells may be a factor of five times more expensive.

037-100 chapter 3 18/11/2004 07:20

Storage in depleted oil and gas reservoirs (including those using EOR and EGR) is perceived tohave a lower leakage risk than aquifers, as the geology of aquifers is often poorly known. It mustbe validated, however, that past exploration operations have not damaged depleted reservoirs andthat the seals of shut-in wells remain intact.

Depleted oil and gas fieldsDepleted oil and gas reservoirs can be filled with CO2. The operation is quite simple, as only aninjection well is needed. Moreover, part of the existing infrastructure may be re-used, which canreduce investment cost. A reservoir may need several injection wells, depending on the field geologyand the rate of injection. The future potential will increase in time as more fields are depleted.Most of the conventional oil and gas production resources are located in the Middle East and theformer Soviet Union (FSU). Using this potential would imply shipping CO2 to these regions. Therefore,the economic potential is smaller than suggested by the figures in Table 3.10.

It is not clear to which extent EOR and EGR can be applied since this will depend on the geologyof a particular field. Also the distinction between EOR and storage in depleted oil fields and EGRand storage in depleted gas fields is not clear-cut. If revenues can be generated from EOR andEGR, such activities would be preferable to simply storing CO2 in depleted oil and gas reservoirs.


Table 3.10

Characteristics of CO2-enhanced fossil fuel production

EOR EGR ECBM

Technology status Proven Speculative Speculative

Cost 5-20 USD/t CO2 5-20 USD/t CO2 10-75 USD/t CO2

Benefits27 0.25-0.5 t oil/t CO2 0.03-0.05 t methane/t CO2 0.08-0.2 t methane/t CO2

15 USD/bbl oil 0.5-3 USD/GJ gas 0.5-3 USD/GJ gas

25-55 USD/ t CO2 1-8 USD/t CO2 2-30 USD/t CO2

Limitations - Oil gravity at least - Depleted gas field - Coal cannot be mined25° API - Local CO2 availability - Sufficient permeability

- Primary and secondary - Maximum depth 2 kmrecovery methods have - Local CO2 availabilitybeen applied

- Limited gas cap

- Oil reservoir at least 600 metres deep

- Local CO2 availability - Global potential All depleted oil fields All depleted gas fields ECBM with net benefits(cumulative)28

2010-2020 35 Gt CO2 80 Gt CO2 20 Gt CO2

2030-2050 100 - 120 Gt CO2 700 - 800 Gt CO2 20 Gt CO2

27. Excludes cost for CO2 injection wells and recovery wells, CO2 recycling and gas preparation. Fuels valued at wellhead price.

28. Note that these potentials only consider storage availability. In practice storage will be limited in the first decades by capture projectexpansion, not by storage capacities. Long-term potentials based on IEA GHG data (Freund and Davison, 2003).

037-100 chapter 3 18/11/2004 07:20


Table 3.11

CO2 EOR projects worldwide

Country Total projects Ongoing projects

USA 85 67

Canada 8 2

Hungary 3 0

Turkey 2 1

Trinidad 5 5

Brazil 1 1

China 1 0

Source: Berge, 2003.

The CENS project

There is considerable interest in the idea of setting up a ‘backbone’ CO2 supply system forthe multiple oil fields in the North Sea that will mature in coming decades. This initiativeis known as the CENS project (CO2 for EOR in the North Sea). The North Sea offers aunique opportunity because of the proximity of large anthropogenic CO2 sources and oilfields. Preliminary estimates suggest that up to 30 Mt CO2 per year could be used forEOR over a period of 15-25 years (Hustad, 2003).

Studies suggest that a combination of IGCC power production with CO2 capture andoffshore EOR in the UK would be a cost-effective strategy, even without credits for CO2emission reduction (Marsh, 2003). The project would have to start soon, however, as thefirst fields reach their EOR stage in 2006. EOR could be postponed for five years (Kaarstad,2003), but not much longer, as the oil platforms would be dismantled or other EOR methodswould be applied. Using EOR at a far later stage would require huge investments andmake such a project uneconomic.

It is important to note that the storage potential in depleted gas fields is much larger than indepleted oil fields. They tend to be bigger reservoirs, and there are more of them. Total capacity isabout 1,000 Gt of CO2, almost 50 years of current global CO2 emissions.

While the storage capacity can be estimated based on the historical quantities of oil and gas produced,the actual storage potential may be reduced as the pressure cannot be brought back to the originalpressure, and parts of the reservoir may be water-flooded. Therefore, these are rough estimates only.

CO2 enhanced oil recoveryUS CO2 EOR oil production amounted to 206 thousand barrels per day in 2003 (Moritis, 2004). Thisequals 31% of total US enhanced oil recovery, or about 0.25% of global oil production. 32 Mt CO2 peryear is used from natural resources and 11 Mt from industrial processes. A few CO2 EOR projects exist

037-100 chapter 3 18/11/2004 07:20

outside the USA (Table 3.11). CO2 EOR has been applied for three decades and should be consideredan established technology. However, this technology has been developed from the viewpoint of oil recovery,not from the viewpoint of CO2 storage. Therefore, some adjustments may be needed for CO2 storage.

EOR can enhance oil production substantially. The additional recovery amounts to 8-15% of thetotal quantity of original oil in place, which increases total recovery by 50% for an average field.Depending on the geology of the oil field and the oil type, enhancement can range from 25-100%. However, CO2 EOR cannot be applied to all fields. An estimate for Norway is that EORcan increase ultimate oil production by 300 million m3 (Mathiassen 2003), which representsabout 10% of production to date and the remaining reserves (IEA 2002a). Given this increase,CO2 EOR can increase the long-term conventional oil supply substantially. Moreover, the EOR revenuescan offset part of the CCS cost.

EOR is limited to oil fields at a depth of more than 600 metres. The oil should also have a gravityof at least 23° API, equivalent to a density of at most 910 kg/m3, which makes this method unsuitablefor heavy oil or oil sands. At least 20-30% of the original oil should be still in place. EOR is limitedto oil fields where primary production (natural oil flood driven by the reservoir pressure) and secondaryproduction methods (water flooding and pumping) have been applied.29 Many oil fields have notyet reached that stage. The occurrence of a large gas cap also limits the effectiveness of CO2 flooding.

Up to temperatures of around 120 °C, CO2 mixes with oil (a so-called miscible flood). At highertemperatures, CO2 replaces the oil (a so-called immiscible flood oil). A miscible flood is moreadvantageous than an immiscible flood, because it results in higher oil recovery factors. Becauseof the physical constraints for CO2 EOR, a detailed field-by-field assessment is required in order toassess its benefits properly. The CO2 storage in case of miscible EOR ranges from 2.4 to 3 tonnesof CO2 per tonne of oil produced. Estimates for storage potentials vary widely from a few Gigatonnes(Gt) to several hundred Gigatonnes of CO2, depending on how many of the cost and geologicalconstraints are considered. The cumulative storage capacity (the total quantity that can be storedover the whole period up to that year) increases with time as EOR can be applied to more depletedoil fields. In a recent study, 420 ‘early opportunities’ for CO2 EOR projects were identified, wherecapture sources and depleted oil fields are within 100 km distance and EOR could start in thecoming years. Assuming around 1 Mt CO2 storage per year per project, this suggests almost 0.5 Gt/yrof storage potential (Bergen et al., 2004).

Worldwide, the potential for CO2 EOR is limited. One reason is that oil fields are not evenly distributedaround the world. The regions with ample oil reserves (Middle East, FSU) are not the regions withimportant point sources of CO2; point sources may be far away from the oil fields (Figure 3.12).CO2 EOR is also not suitable for all oil fields. Moreover, it competes with other EOR technologies(see Table 3.12). It depends on the reservoir and local supply conditions as to whether CO2 floodingreally is the best option from an oil recovery perspective, especially for heavy oil (i.e. oil of adensity of more than 0.9 t/m3) and for oil fields with a significant gas cap. The share of CO2 EORin total EOR has been expanding rapidly during the last two decades. If CO2 were available atlow cost, CO2 EOR could be applied to the majority of the world’s oil fields.

CO2 EOR investment costs have dropped from more than 1 million USD per site in the 1980s toless than half of that today. Project costs vary depending on field size, pattern spacing, locationand existing facilities, but in general, total operating expenses (exclusive of CO2 cost) range from2-3 USD per barrel (bbl). Costs can be split into capital costs (about 0.8 USD/bbl), operating costs(2.7 USD/bbl), royalties taxes and insurance (3.6 USD/bbl) and CO2 costs (3.25 USD/bbl) (KinderMorgan, 2002). Typically, to flood a field with CO2, the field should have more than 5 million barrels


29. There are examples of CO2 EOR being applied as secondary oil production technology.

037-100 chapter 3 18/11/2004 07:20


Table

3.1

2

Scre

enin

g c

rite

ria

for

enh

ance

d o

il re

cove

ry m

eth

ods

EOR

met

hod

°API

IVi

scos

ity

(cp)

Com

posi

tion

Oil

satu

rati

onFo

rmat

ion

type

Net

thi

ckne

ssPe

rmea

bilit

yD

epth

TCo

st(%

PV)

(m)

(md)

(m)

(°C)

(USD

/bb

l)

N2

(&flu

e ga

s)>3

5/48

<0.4

/0.

2H

igh

% C

1-C7

>40/

75Sa

ndst

one/

Thin

unl

ess

->2

,000

-Ca

rbon

ate

dipp

ing

Hyd

roca

rbon

>23/

41<3

/0.

5H

igh

% C

2-C7

>30/

80Sa

ndst

one/

Thin

unl

ess

->1

,350

-

Carb

onat

edi

ppin

g

CO2

>22/

36<1

0/1.

5H

igh

% C

5-C1

2>2

0/55

Sand

ston

e/-

->6

0012

0307-

3031

Carb

onat

e

Mic

ella

r/

>20/

35<3

5/13

Ligh

t, >3

5/53

Sand

ston

e-

>10/

450

<3,0

00/

<95/

258-

12

inte

rmed

iate

1,10

0

Poly

mer

flo

odin

g>1

5/ <

40<1

50/

>10

->7

0/80

Sand

ston

e-

>10/

800

<3,0

00<9

5/60

5-10

Com

bust

ion

>10/

16<5

,000

/-

>50/

72H

igh

poro

sity

>3

0>50

<4,0

00/

>40/

553-

6

poly

mer

, Alk

alin

e/1,

200

sand

/sa

ndst

one

1,20

0

poly

mer

Alk

alin

e

flood

ing

Stea

m>8

/13

.5<2

00,0

00/

->4

0/66

Hig

h po

rosi

ty

>6>2

00<1

,500

/-

3-6

4,70

0sa

nd/

sand

ston

e50

0

Not

e: T

he s

econ

d fig

ure

indi

cate

s cu

rrent

ave

rage

con

ditio

ns. P

V =

Pore

Vol

ume.

Sour

ce: G

reen

and

Will

hite

, 199

8, p

.9.

30. F

or m

isci

ble

CO2

flood

s.

31. L

ower

end

ass

umes

CO 2

is a

vaila

ble

for f

ree,

hig

her e

nd in

clud

es C

O 2co

st.

037-100 chapter 3 18/11/2004 07:20


Figure 3.12

Known conventional petroleum reserves of the world by region

Key point: The greatest potential for CO2-EOR is in the Middle East and the FSU

0

50

100

150

200

250

300

350

400

450

500

Undiscovered

Remaining

Production to date

Oth

ers

Indo

nesiaU

K

Nor

way

China

Braz

il

Vene

zuela

Liby

a

Alge

ria

Nig

eria

Mex

ico

USA

Kaza

khstan

Russia

Neu

tral z

one

Qat

ar

Kuw

ait

UAEIra

nIra

q

Saud

i Ara

bia

bln

BO

E

Note: Excludes oil- and tar sands. Bln BOE = billion barrels of oil equivalent.

Source: IEA 2002a, p. 97

of original oil in place and more than 10 producing wells (Kinder Morgan, 2002). With EOR, totalproduction costs (excluding CO2 costs) are approximately 7 USD/bbl oil (about 50 USD/t oil). Ata wellhead oil price of 15 USD/bbl and assuming an injection rate of 2.5 t CO2 per tonne oil, therevenues amount to 25 USD/t CO2, if the CO2 is available for free. Note that this assumes a highoil recovery per tonne of CO2. Oil revenues would be lower for most other fields.

Injection wells constitute the bulk of the capital costs for storage. In the case of depleted oil andgas wells it is recommended to drill new CO2 injection wells, because there is a threat of a blow-out when old and possibly damaged production wells are used (Over et al., 1999).

The EOR impacts on oil supply can be substantial. Taking the case of Norway again, it is estimatedthat CO2 EOR can result in an additional oil production of 260 to 300 million m3. This equals3.3 to 3.8 percentage points of the original oil in place. Given an original recovery factor of44 percentage points, this represents an increase of 7.4% to 8.5% (Mathiassen, 2003). This estimateis based on a detailed field-by-field analysis, taking the feasibility of EOR for various fields intoaccount. The recovery factor is set at 4% to 8%, roughly half the US experience, based on oilcompany feasibility studies for four North Sea fields. This suggests that the oil recovery potentialmay be higher than projected.

CO2 enhanced gas recoveryUsing CO2 for enhanced gas recovery (EGR) is a speculative method for repressurizing depletedgas fields that can be applied to certain fields when 80-90% of the gas has been produced. Although

037-100 chapter 3 18/11/2004 07:20

target reservoirs for CO2 sequestration are depleted in methane with pressures as low as 20-50bars, they are not devoid of methane. Additional methane can be recovered by injecting CO2 usingEGR (Oldenburg and Benson, 2001). The injected CO2 will flow in the reservoir due to pressureand gravitational effects. Regardless of phase (gaseous, liquid or supercritical) CO2 is notably denserthan CH4 at all relevant pressures and temperatures and will tend to flow downwards, displacingthe native CH4 gas and repressurizing the reservoir (Oldenburg and Benson, 2001). If CO2 is injectedat the bottom of a gas reservoir, it will ‘chase’ the gas toward the top where it can be produced.Not every gas field is suitable for CO2 injection, however.

CO2 EGR has not yet been applied anywhere in the world. Opinions are divided on whether thistechnology is feasible for most gas fields. It will depend on factors such as the time needed for theCO2 to reach gas production wells. Modelling studies suggest that after 10 years the gas producedwould still contain only 10% CO2 by mass (Oldenburg and Benson, 2001). Given an initialpressure of 120 bar, another 5-15% of the initial gas in place could be recovered using EGR. Theactual percentage depends on the geology of the gas field, and on the operator’s selection of thepercentage of CO2 in the gas produced at which the EGR operation is closed down (Clemens andWit, 2001).

About 1.8 GJ of gas could be recovered per tonne of CO2 stored, if a whole reservoir was filled withCO2 up to its original pressure.32 Modelling studies for the Netherlands suggest that a coal-firedIGCC in combination with CO2 removal and injection in a depleted gas field would be an economicoption (Clemens and Wit, 2001). The potential for CO2 use for EGR might be larger than for EOR(see Table 3.10). Per tonne of CO2, the EGR revenues are substantially lower than for EOR.

CO2 enhanced coal-bed methane recoveryCO2 enhanced coal-bed methane (ECBM) is a speculative method for methane (coal gas) recoveryfrom coal seams. While conventional coal-bed methane recovery may achieve 40-50% recovery(close to the wells), the recovery increases to 90-100% in the case of ECBM. ECBM is limited tocoal seams that will not be mined.33

ECBM can only be applied to coal seams of sufficient permeability. Because of the increasingpressure, the CO2 adsorption increases from 2 mole per mole methane at 700 metres, up to 5 moleper mole at 1,500 metres (Bergen et al., 2000). The coal reserve should not be deeper than2,000 metres because the increasing temperature limits the methane content of the coal and theincreasing pressure at greater depth reduces the coal seam permeability. The methane content ofdeep coal seams can vary from 5-25 m3/t coal and the thickness of the coal seams varies, so theECBM potential per well and the CO2 storage costs will vary by a factor of five or more. It is worthnoting that the most attractive option from a methane recovery perspective (shallow coal reserveswith thick coal layers) is the least attractive one from a CO2 storage perspective and from a futurecoal mining perspective.

The following criteria must be met when screening coal reservoirs for ECBM (Stevens, 1998). Onlya small fraction of all coal seams meet such criteria:

● A homogeneous reservoir, laterally continuous and vertically isolated from surrounding strata;

● Minimally faulted and folded;

● At least 1-5 millidarcies (mD) permeability. Most coal seams are much less permeable;


32. 1.8 = 16/44x50x0.1

33. This constitutes a major source of uncertainty, because it depends on future mining technology and energy demand trends.

037-100 chapter 3 18/11/2004 07:20

● High methane content;

● Stratigraphically concentrated coal seams are preferred over multiple thin seams;

● A possibility to use or export methane (pipeline) and CO2 availability (local power plant, industryor pipeline).

The worldwide potential for CO2 sequestration in deep unminable coal seams has been estimatedat 148 Gt CO2. An analysis of representative CO2 ECBM projects indicates that 5-15 Gt of CO2 couldconceivably be sequestered at a net profit, while about 60 Gt of sequestration capacity may beavailable at cost of less than 50 USD/t CO2, not including the cost of capture and transportation(IEA GHG Programme, 1998; Gale, 2004). Assuming that 2 moles of CO2 replace one mole ofmethane, 10 Gt of CO2 would equal 90 EJ of natural gas, which equals the current consumptionof one year.

Note that these potentials depend critically on the assumptions regarding coal permeability, thecosts for enhancing coal seam permeability and the costs for injection wells. One of the majorproblems concerning widespread use of CO2 for ECBM is the variable, and often low, permeabilityof the coal. Furthermore, coal tends to swell in contact with CO2 which reduces permeability. Theonly successful operation to date, the Allison unit in the San Juan Basin in the USA, is notrepresentative because of very specific conditions. Low permeability can, in some cases, be overcomeby fracturing the formation.

The cost of wells rises exponentially with the depth of the coal seam (compare Figure 3.11). Alsodeviated drilling is 70% more expensive than vertical drilling (Bergen et al., 2003). Well costs areimportant with ECBM because a high well density is needed. As Figure 3.13 shows, well costs canrepresent three- quarters of total project costs. As a result, costs will vary on a project-by-projectbasis. Moreover, drilling a large number of wells could face opposition from land owners. Another


Figure 3.13

Cost breakdown of a proposed ECBM project in the Netherlands

Key point: Well costs represent two-thirds of total ECBM investment costs

Production well investment44%

Injection well investment22%

Production well O&M4%

Injection well O&M8%

CO2 supply15%

Water storage5%

CBM treatment2%

Note: Annuity for investments is 13%. Deviated injector wells because of siting constraints. CO2 supply costs exclude capture, butinclude pressurization and transportation.

Source: Bergen et al., 2003.

037-100 chapter 3 18/11/2004 07:20

general problem for coal-bed methane projects is the necessary coal dewatering, which results inlowering of the groundwater levels (for shallow reservoirs) or production of brackish/salty water(for deep reservoirs).

So far, CO2-ECBM pilot projects have been undertaken in the USA and Canada and a third pilotproject in Poland is well underway. Results so far do not show conclusively that CO2 enhancescoal-bed methane production. The energy penalties and costs of ECBM are still unclear (C3 views,2003). A case study for the Netherlands suggests transport and storage costs in the range of 55-75 USD/t of CO2 – excluding the cost of CO2 capture – at a natural gas price of 3.5 USD/GJ(Bergen et al., 2003).

In conclusion, CO2-ECBM technology is at an early stage of technical development and its prospectsremain uncertain. New demonstration projects currently underway should provide valuableinformation on the technology and allow a decision to be made within a few years on whether thiscan be regarded as a safe and environmentally-acceptable mitigation option. In most cases, revenueswill be limited compared to the additional costs incurred. Obviously, high cost storage makes littlesense if low-cost storage in depleted oil and gas fields and aquifers is an option.

Storage in deep saline aquifers

An aquifer is a layer of sedimentary rocks saturated with water and from which water can be producedthrough pumping, or into which fluids can be injected. Sandstone and carbonate rocks are usuallyaquifers. However, while most pore and fracture space in rocks are filled with water, only sedimentaryrocks, such as sandstones and carbonates, have sufficient porosity to be considered for CO2 storage.Crystalline and metamorphic rocks, such as granite, do not have the porosity necessary for CO2 storage.In addition, fracturing usually present in the latter creates potential leakage paths.

An aquitard is a layer of rock from which water cannot be produced, but that has enough porosityand allows the flow of water on a geological timescale. Shales usually constitute aquitards. Anaquiclude is a layer of rock that has almost no porosity and does not allow the flow of water. Saltand anhydrite beds are aquicludes. Water in aquifers deep below the ground in sedimentarybasins is confined by overlying and underlying aquitards and/or aquicludes, usually has a highcontent of dissolved solids (brackish water and brine) and may have been there for millions of years.This water is unsuitable for human consumption. Because of the confined character of these aquifersand the lack of alternative applications, they have been proposed as locations for CO2 storage.

Open and confined aquifers exist. The former type has no natural barriers to water flow, and theremay be a natural circulation at a very low speed. Closed aquifers have no such circulation. Therefore,they might be better suited for CO2 storage. Geological CO2 sequestration in divergent basins(such as the foreland basins east of the Rocky Mountains and the Andes, the Michigan basin andthe North Sea) is much safer than in convergent basins (California, Japan and New Zealand) becauseof the tectonic stability and general lack of significant hazardous events. Old continental coreareas (e.g., the Canadian and Brazilian shields) and mountain forming areas do not possess therock characteristics necessary for CO2 sequestration (Bachu, 2000). Sedimentary basins can befurther subdivided in a number of criteria (Bachu, 2003). Based on such analysis, not every basinis suited for CO2 storage.

Storage in aquifers is currently being studied in the Statoil CO2 storage project in the North Sea. CO2

is separated from natural gas produced from the Sleipner field, and stored in the Utsira aquifer belowthe gas field. The project has been storing 1 Mt CO2 per year since late 1996. So far, results suggestthat there is no leakage and CO2 storage is technically feasible. However, there is still considerable


037-100 chapter 3 18/11/2004 07:20

uncertainty regarding storage potential. The main uncertainty is to what extent the aquifer pore volumecan be filled with CO2. Calculations from the beginning of the 1990s suggest that 2% of the aquifervolume can be filled with CO2 (Meer 1992), but more recent estimates suggest between 13% and68% (Holt et al., 1995). The higher the average storage efficiency, the fewer the number of wells thatwill be required, the lower the storage costs and the higher the storage potential.

CO2 injected in deep saline aquifers is trapped and stored by several mechanisms: 1) in its freephase as a plume at the top of the aquifer and in stratigraphic and structural traps (similar to oiland gas accumulations); 2) as bubbles that are trapped in the pore space after passing of a plume;34 3)dissolved in aquifer water; and 4) as a precipitated carbonate mineral as a result of geochemicalreactions between the CO2 and aquifer water and rocks. Numerical studies have shown that,during the period of injection, up to 29% of the CO2 would dissolve in the brine (Bachu, 2000).As CO2 has a lower density than the brine, the remainder would float on top of the brine andaccumulate below the cap rock. During later periods, part of this CO2 may dissolve in the brine orreact with the aquifer rock matrix. Dissolution would continue after injection has ceased so that,over a period of a thousand years or more, the entire plume of CO2 would probably dissolve.

Geochemical reaction to permanently sequester CO2 would take several thousand years to have asignificant effect. Where there is no stratigraphic or structural trap, the CO2 would flow and spreadover a large area below the aquifer cap rock. Modelling studies suggest a spread of tens or hundredsof square kilometers, depending on aquifer properties such as thickness, porosity and permeability(Saripalli and McGrail, 2002). This also depends, however, on the topography of the cap rock andthe volume injected.

Generally, modelling studies have shown that, depending on aquifer characteristics and injectionrate, a plume of CO2 may spread between five and twelve kilometres from the injection well overa period of 1,000 years. Other studies suggest the plume would dissolve entirely. Such a large areawould complicate the monitoring and verification of leakage, but the area needed would vary bycase. The lower the initial CO2 saturation of the brine, the smaller the area needed, as more CO2

would dissolve in the brine.35 This relationship could be used as part of aquifer selection criteria.


Table 3.13

Recent estimates of CO2 storage potentials in deep saline aquifers

(Gt CO2)

Alberta (Canada) 4,000

USA 5-500

Western Europe 800

Worldwide 100-10,000

Note: Includes offshore aquifers.

Source: IEA GHG 2001; Bruant et al., 2002; Christensen, 2003; Bachu and Adams, 2003.

34. This process, also called imbibition trapping or residual gas trapping has received a lot of attention recently, with claims that it could trap 5-25% of the CO2 injected. However, these estimates are based on model observations, calibrated with models for natural gasproduction reservoirs. A fundamental difference is that CO2 dissolves in water, while natural gas does not dissolve. Therefore diffusionmay reduce this pore phase trapping on the longer term, and it might therefore not contribute to the long-term permanence of CO2

storage35. It may be possible to mix CO2 with brine before injection and inject the CO2 in its dissolved state rather than as gas. While thisoption is speculative, it would reduce the leakage risk..

037-100 chapter 3 18/11/2004 07:20

The temperature profiles in underground sediments differ by location, because of variations ingeothermal gradients and in surface temperatures. As a consequence, the state of CO2 undergroundwill vary, as will the density at a given pressure (Bachu, 2000). This affects both the storage potentialper unit of surface and the relevance of leakage mechanisms. Aquifer CO2 storage estimates areshown in Table 3.13. Estimates vary widely. For example, estimates from the USA of several hundredGt CO2 are almost a decade old and are probably too low.

More detailed assessments for the US mid-western Mt. Simon aquifer alone indicate 115 to 655 GtCO2 of storage potential (Gupta et al., 2002) although storage potential of several thousandGigatonnes is more likely. In conclusion, significant storage potentials exist, but they are not spreadevenly across and within all regions. Ongoing studies are attempting to match potential capturesites with storage sites. Over a timescale of decades, it may be that power plants and industrial plantsare relocated to sites where suitable transportation and storage infrastructure exists.

This conclusion is valid on a global scale, but on a regional scale limitations may exist. A simpleback-of-the-envelope calculation clarifies the issues involved. A 500 MW coal-fired power plant wouldhave to store about 3 Mt CO2 per year. The CO2 may be stored in an aquifer at a density of 0.5 t/m3.Assuming an effective CO2 layer density of 1 metre,36 six square kilometres of aquifer are used forstorage per year. During a power plant lifespan of 40 years, an area of 240 square kilometreswould be needed, which equals an area of 15 by 15 kilometres. Some 16 Gt CO2 of storage peryear – a huge amount – implies an area for underground storage of 200 by 200 kilometres peryear, the size of a country such as the Netherlands. This rough estimate indicates that CO2 storagewould require geo-engineering on a global scale.


36. If the sediment porosity is 30%, that means the top 3 metres of the aquifer are filled with CO2.

Figure 3.14

The cost structure of Norway’s Snohvit aquifer storage project

Key point: Pipeline and CO2 compressor costs account for three-quarters of Snohvit’s investment costs

Well completion5%

Pipeline 8", 160 km38%

Control umbilical (sub sea)6%

Sub sea well frame6%

CO2 compressor train37%

Offshore CO2 well8%

Source: Audus, 2003.

037-100 chapter 3 18/11/2004 07:20

The cost of the Sleipner project for CO2 compression and injection amounted to 80 million USD,equal to an investment cost of 80 USD/t of CO2. The investment costs for the Snohvit project(compression, transportation and injection) will amount to 191 million USD, equal to an investmentcost of 275 USD/t CO2 (Audus, 2003). Clearly these cost levels are higher than the values usedfor regular CCS assessment studies. The higher costs may be explained by extreme situations(North Sea offshore and Arctic, respectively), and by the fact that these are ‘first of a kind’ facilities.Yet, compressors and pipelines constitute the bulk of the cost (Figure 3.14), and these should beconsidered well established pieces of equipment, for which the learning potentials are limited.Therefore, careful case-by-case cost evaluation is needed, and economies of scale will be essential,in determining the most cost-effective and environmentally sound options for CO2 storage.

Other storage optionsA number of other storage options have been proposed. Only limestone ponds, surface mineralizationand oceanic storage will be discussed here.

The concept of limestone ponds combines capture and storage. Limestone is dissolved in water ina pond. Flue gas is bubbled through this pond. The CO2 in the flue gas bubbles reacts with thelimestone. The carbonate solution is dissolved in sea water. There have been preliminary costestimates of 21 USD/t CO2 for the total of capture and storage (Sarv and Downs, 2002). Thisprocess has not been proven on a pilot scale. Most experts claim that it is impossible to producebubbles that are sufficiently small (CO2 transportation into the solution is the limiting factor), andthe size of the ponds would be prohibitive. Given its speculative character, this option has notbeen considered in more detail.

The concept of surface mineralization is based on the reaction of ground magnesium and calciumsilicate rock with CO2 into carbonates. The volumes of material involved are significant. A 500 MWcoal-fired power plant would produce about 30 kt of magnesium carbonates a day (about 1,000 truckloads; Goldberg et al., 2001). The process has been tested on a laboratory scale. Certain types ofperidotites and sepentinite would be the preferred rocks, containing 40-50 weight % MgO andCaO. These rocks occur worldwide. Binding one tonne of CO2 would require 0.9 t of MgO, andgenerate 2.8 t of waste (Lackner et al., 1997). These rocks, which are not sedimentary, occur worldwidein specific areas (i.e., not in sedimentary basins where energy is usually being produced). For example,olivine and serpentine is found in North America on the east and west coast, while oil, gas andcoal are produced mostly between the Appalachian and Rocky Mountains.

Advocates of this process argue that it is exothermic and, therefore, that the energy requirementswould be negligible. Moreover, the olivine and serpentine starting materials are abundant. The mostimportant hurdle from an engineering perspective is the reaction kinetics. So far, no process designhas been proposed that results in realistic reaction times (Herzog, 2002). It should also be notedthat large-scale application of this process would create large amounts of solid waste that requirefurther processing. Finally, the process is not cheap. The goal is to reach storage costs of 30 USD/tCO2 (excluding CO2 capture and transportation), (Lackner, 1997). Transportation of olivine andserpentine to CO2 sources, or of CO2 to quarries where the former could be mined would make theprocess prohibitively expensive. As a result, this process would only be attractive if other storageoptions are unavailable (e.g., because of environmental concerns). Mineralization has been consideredin the ETP model.

Oceanic storage of CO2 is the most controversial option. There are two types of storage: dissolutionin seawater and storage of CO2 hydrates or liquid CO2 at depths of more than 4,000 metres. Mosttechnologies for deepwater storage are established technologies. Little is known about the impact


037-100 chapter 3 18/11/2004 07:20

increasing CO2 concentrations would have on the oceanic ecosystem. Pilot projects in Hawaii andNorway were cancelled because of protests from environmentalists. While oceanic storage is notcritical for Western countries it may emerge as a key alternative for Japan because of the country’slimited underground storage potential. However, this is a country where the sustainable use ofoceanic resources is a sensitive issue. Wide acceptance of environmentally-friendly oceanic storagesystems would be a key requirement for the large-scale application of this option.

With regard to oceanic storage, model calculations suggest that 80% to 90% of the CO2 wouldremain in the ocean after a period of 500 years, if the CO2 was injected at a depth of over3,000 metres. For a depth of 1,000 metres, 30-80% would remain (Caldeira, 2002). The differencebetween the lower and the higher end of the range is the CO2 that is released into the atmosphereand reabsorbed. The lower end of the range should be used for proper comparison of the efficiencywith other options. These figures suggest that permanence is less of an issue if the CO2 is injectedat sufficient depth. In this case it is probably not the permanence but the direct environmentalimpacts on the sea ecosystem through a change in water acidity that are the main obstacle. If thecarbon acidity is not neutralized with limestone or some other buffer, the addition of thousands ofgigatonnes of carbon to the ocean will produce significant perturbations to ocean chemistry on alarge scale. It is unclear at this time how best to monitor the health of broad reaches of the oceaninterior, when so little is understood about these ecosystems. Again, more research is required tobetter understand deep-sea biota and its response to added CO2 (Caldeira, 2002). In addition,international legal issues would pose a significant barrier to large-scale implementation of oceanstorage.

CO2 Storage: Permanence and Monitoring

The idea of storing CO2 in geologic formations immediately raises questions about storagepermanence, the environmental risks involved and necessary monitoring. Certain potential storagesites may not leak at all, while others may do so at an unforeseen rate. At the moment, insufficientinformation is available to quantify leakage from CO2 storage sites. It is possible, however, to quantifyupper limits for leakage and to draw conclusions from these theoretical limits and the experimentalinformation available so far.

A strict requirement for a zero leakage rate would impose excessively stringent conditions on storageselection procedures and result in a waste of a valuable resource, i.e., potential CO2 storage sites.Certain leakage rates can be accepted and permitted. It has to be emphasized, however, thatselection procedures should effectively eliminate sites with a risk of sudden releases of bulk CO2

due to geological imperfections and tectonic moves.

There are two types of risk associated with leakages of CO2: local, site specific, affecting health, safetyand environment, and global, resulting from a return of stored CO2 to the atmosphere. The majorityof constraints imposed on storage permanence and also quality of monitoring will probably resultfrom the first type (Keith, Pacala 2004). A local risk resulting from leaking CO2 is a very site-specificissue, however, and will not be covered in the following discussion on maximum leakage rates.

Taking the global risk under consideration, the minimum storage permanence time depends onfuture emissions. The total quantity of fossil fuels in place (about 5.67 PtC remaining) puts an upperbound on required storage time. Oil reserves are probably most limited, followed by gas and coal.Coal reserves are very large and could last for hundreds of years. If CO2 concentrations should notrise above 450 ppm this would imply a retention time of 7,000 years (Zweigel and Lindeberg, 2003).


037-100 chapter 3 18/11/2004 07:20

On the other side of the scale, non-fossil power generation may become dominant in the secondhalf of the 21st century. If fossil fuels are eliminated by 2100, then CO2 storage for 100 years wouldsuffice, according to this author. However, if large quantities of CO2 are stored during this century,such a short retention time (or such a high leakage rate) will be hardly compatible with stabilizingCO2 concentrations at any level, as stabilization of CO2 concentrations will require near-eliminationof net CO2 emissions. Any storage time shorter than 100 years is questionable in all scenarios. Ingeological terms, these are very short periods. Oil and gas have been buried for millions of years,indicating that such favourable storage conditions are not uncommon.

A retention time between 100 and 2,000 years means the maximum acceptable leakage rate canbe somewhere between 0.01% and 1% per year. However the more optimistic scenario is due toan assumption of heterogeneity among reservoirs. The emissions from leaky reservoirs are re-injectedinto more average reservoirs with much lower seepage rates, thereby reducing the average seepagerate over time (Torvanger et al., 2004). Other studies have found that leakage rates of up to0.1% per year allow for an effective storage policy (Hepple and Benson, 2003; Pacala, 2003).There are two important issues that need to be mentioned here:

● Maximum allowable leakage rates will set the upper bound on CO2 losses in permitting andaccounting procedures although this does not mean that the research community expects suchleakages which, in reality, should be many times smaller, if any;

● Because of public perception issues, a maximum leakage rate considered in a site selectionprocess will likely have to be one order of magnitude smaller than that resulting from thecalculations.

One of the main elements of the site selection procedure is an assessment of faults and fracturesthat can compromise the cap rock strata (Friedmann and Nummedal, 2003). Usually, depleted oiland gas reservoirs are well characterized, so these imperfections can be identified without high cost.For aquifers, such data is not readily available and, when available, is not at the same level ofdetail and resolution. A costly suitability study may be needed.

Model studies suggest that a fracture 8 km from the injection well would result in preliminaryleakage of CO2 after 250 years and 10-20% leakage over the succeeding 2,000 years, equal toless than 0.01% per year (Lindeberg, 1997). Anthropogenic damage of the cap rock due to abandonedoil and gas exploration and production wells may cause additional leakage (Celia and Bachu, 2003).In some regions with a well-developed oil and gas industry, more than five wells occur per squarekilometre. Most of the abandoned wells have been sealed. However, CO2 reacts with the cement,often used for the seal, and can result in leakage. Also small gaps may exist between the well plugand the well casing. Leaking CO2 may dissolve in other aquifers above the storage aquifer, thuspreventing an emission to the atmosphere. So far, it is not clear whether this leakage mechanismposes a serious problem or not.

On the other hand several natural processes could enhance the permanence of storage (see previousdiscussion on storage in deep saline aquifers). The dissolution of CO2 in the aquifer water is a keyprocess. The solubility of CO2 in 1 mole/l (M) brine reaches a maximum at 41-48 kg/m3 below600 metres depth. Increasing the salinity to 4 M decreases the maximum solubility to around 24-29 kg/m3. Geochemical reactions can increase this solubility by 20%. This dissolution is kineticallylimited and only takes place on a timescale of hundreds of years. Therefore, storage in watercontaining reservoirs (both aquifers and depleted oil and gas fields) should be interpreted as storageof a CO2 ‘bubble’ on top of a water layer (Rigg, 2001).

The dissolved CO2 results in a type of ‘sparkling mineral water’, which occurs naturally in manyplaces. For example, in many geothermal energy projects natural CO2 from deep saline aquifers is


037-100 chapter 3 18/11/2004 07:20

released into the atmosphere. In the high emission scenario, dissolution would reduce the minimumretention time for the CO2 ‘bubble’ from 7,000 to 2,000 years (Zweigel and Lindeberg, 2003).Chemical reactions with the rock would further reduce the retention time. In certain cases, physicaladsorption can fix the CO2, e.g., in coal and so-called residual trapping mechanism,. Also, leakingCO2 would change from a supercritical state into a subcritical state. This phase change wouldresult in strong cooling, followed by formation of solid water ice and CO2 hydrate. Modelling studiessuggest that this process could inhibit leakage for hundreds of years (Pruess, 2003). ‘Fixing’ CO2

through different mechanisms increases security of storage and, in certain cases, may result in asoftening of monitoring requirements with time.

In recent decades there has been significant progress in the monitoring of underground oil andgas reservoirs with an improvement in 3-D and 4-D seismic methods. These methods can also beapplied for monitoring of CO2 in deep aquifers. However, many key aspects of fault geometry arebelow the resolution of existing seismic tools (Friedmann and Nummedal, 2003). A wide range ofadditional monitoring techniques may be applied, but these often require costly drilling (Vendrig etal., 2003). The Sleipner CO2 injection has been monitored in the Saline Aquifer CO2 Storage (SACS)programme. The cost of that programme amounted to 4.5 million USD, but this should be consideredas a R&D programme with high costs; the cost for routine monitoring will be considerably lower.

On-land, 3D seismic may cost between 6,000 and 10,000 USD/km2. If 25 km2 is to be monitored,the seismic cost would amount to 150,000-250,000 USD. Assuming 10 Mt storage, this wouldamount to 0.15-0.25 USD/t CO2. If seismic monitoring is undertaken at five or ten-year intervals,the cost may be of secondary importance. Other analyses estimate slightly higher total undiscountedstorage (pre-operational, operational and closure) monitoring cost of 0.19-0.31 USD/t CO2, anddiscounted monitoring cost of 0.05 to 0.10 USD/t CO2 (Benson, 2004).

Simpler monitoring methods could be applied, such as surface measurements of CO2 concentrations.It is possible to measure a flux resulting from a 0.01% leakage per year and differentiate it frombackground emission (Benson, 2004). This would allow for verification of storage permanence. Itis also sufficient for early recognition of certain leaky storage sites, and planning a remediationaction. Various new methods are being developed, e.g., in the framework of the CO2 Capture Project.A list of technology gaps includes instruments capable of measuring CO2-levels close to thebackground and to distinguish between CO2 from natural processes and that from storage. Improvedmathematical models may also contribute to a better understanding of long-term storage permanence.

Calls for better monitoring methods should not be taken as an endorsement of a ‘some is good,more is better’ view of monitoring (Keith, 2004). One of goals of storage demonstration projectsis to define appropriate levels of monitoring for each particular type of storage site.

Out of many natural and industrial analogues, underground natural gas storage operations canprovide very useful information. Underground storage of natural gas is widely applied, e.g., in theNetherlands and the USA. In the USA, 119 Mt of natural gas was stored underground in 2002.Underground gas storage has been practiced for more than 90 years without problems, whichsuggests that underground CO2 storage may be feasible as well.

Well mechanical flaws and abandoned wells have been the most common cause of leaks inunderground gas storage facilities. Generally, such problems are fixed by repairing or reconditioningof the wells. For gas projects in the USA, overpressures of up to 17 kPa/m of depth have been used(Lippmann and Benson, 2002). At a depth of 800m, this would amount to an overpressure of 13,600kPa, or 136 bar. Such high additional pressures are not proposed for CO2 storage, which suggests caprock fracture may be less of a problem. While such comparisons provide some guidance, differences


037-100 chapter 3 18/11/2004 07:20

exist. The storage of natural gas is different from CO2 storage because natural gas does not chemicallyreact with the reservoir rock, and it is stored and removed periodically (Vendrig et al., 2003).

Another commercial-scale analogue to CO2 storage in geological media exists in North Americawhere, since 1990, acid gas (a mixture of hydrogen sulfide H2S and CO2) has been injected intodeep aquifers and depleted oil and gas reservoirs at more than 60 locations (40 of which havebeen in the Alberta basin in western Canada). Acid gas is produced through desulphurization ofproduced sour gas (natural gas that contains H2S). H2S is captured using a chemical absorptionprocess. In this process significant amounts of CO2 are also captured. The purpose of acid-gasinjection operations is to dispose of the H2S. Significant amounts of CO2 are injected at the sametime because it is costly to separate the two gases. At end-2002, 39 acid gas injection projectswere in progress in North America. The cumulative injection of CO2 in all sites exceeds 1 Mt. Inthe 13 years since acid gas injection started in western Canada, no safety incidents have beenreported to the regulatory agencies (Bachu et al., 2003). Plans are currently underway to applythis technology in Kazakhstan, the Middle East (Iran) and North Africa.

Production of Chemicals and Fuels from CO2

At first sight, the production of transportation fuels from CO2 does not seem a viable strategybecause the energy of a fuel is released by its conversion into CO2. The process only makes sensefrom an energy perspective where in one location a surplus exists of CO2-free energy (either ofnuclear or of renewable origin), while in another location a demand exists for fuels. CO2 is shippedfrom one region to the other, while hydrocarbon fuels that are produced from this CO2 are shippedin the other direction. The rationale would be that transportation of CO2-free energy carriers (electricity,hydrogen) is costly. Also, for certain end-use sectors, notably the transportation sector, the introductionof CO2-free energy is problematic.

If CO2-free energy can be converted into a hydrocarbon energy carrier that can be transported andused at low cost, it would allow for the introduction of CO2-free energy in markets without a localCO2-free supply. In the transportation sector, hydrocarbons are the preferred fuel because of theirhigh energy-to-weight and energy-to-volume ratio. Methanol and DiMethylEther (DME) are prominentcandidates, because these fuels can be used in current combustion engines and in a reformer/fuelcell combination that may be the long-term propulsion system of choice.

Transportation and on-board storage of methanol and DME is considerably cheaper than hydrogenor electricity storage, and these fuels can be used in existing combustion engines. However, arenewable carbon source is required for CO2-free methanol and DME. Biomass is the only renewablecarbon source. CO2 that is recycled from flue gases can be another carbon source that results in asignificant overall emission reduction.

This strategy is being looked at in Japan. The synthesis of methanol via CO2 hydrogenation isconsidered one of the most promising processes for the fixation and utilization of CO2 (Takeuchiet al., 1999). The production of the hydrogen for the hydrogenation process requires significantamounts of energy. Hydrogen production accounts for 93% of the total energy required and CO2

separation and liquefaction account for 6%. Total energy requirement is 28 GJ of electricity pertonne of methanol. In the case of a CO2-free electricity source, the methanol constitutes a CO2-freeenergy carrier. If this methanol is used to produce transportation fuels, it results in 3.14 tonnes ofCO2 emission reduction per GJ of oil transportation fuel that is substituted. In the case of an average


037-100 chapter 3 18/11/2004 07:20

emission for electricity production of 0.1 CO2/GJ, the net balance is an emission of 1.4 t CO2/tmethanol, and there is no emission reduction per GJ of oil transportation fuel that is substituted.

Obviously, the use of CO2 feedstock would be a costly strategy for GHG emission reduction, becauseCO2-free electricity is a costly energy source. However, trade among regions with special characteristics(e.g., Australia with ample solar/wind resources and Japan with limited renewables potential andlacking indigenous oil resources) could make such a combination a viable alternative.

Overview of CCS Costs

CCS cost figures have been discussed throughout this chapter and the impact of methodologicalchoices on the cost per tonne of CO2 was discussed. It is clear that cost figures should not be appliedindiscriminately. However, it is useful to give an overview of the cost range, and the main factorsthat determine these costs.

In most cases, the bulk of costs are for CO2 capture and pressurization. There are a few exceptionalcases where CO2 is already separated from gas flows for other reasons. If this is the case, the onlycosts are for compression. In most cases, CO2 must be separated from a gas stream. It depends onthe CO2 concentration and process design to determine which capture technology is best. Generallyspeaking, the capture costs per tonne of CO2 are lower for coal-fired processes than for gas-firedprocesses, as CO2 concentrations are higher. Improving technology can reduce capture costssubstantially, to 5-30 USD/t CO2 (Table 3.2). Costs could decline to 10-25 USD/t CO2 for coal-fired power plants and to around 25-30 USD/t CO2 for gas-fired plants; they could be even lowerfor biomass fired processes. The gap between capture and abatement cost narrows as the energyefficiency penalty for CO2 capture decreases.

CO2 transportation costs depend on volume, distance, population density (land prices), soil typeand other factors. With optimistic assumptions, these costs may only amount to 2-10 USD per tonne.Low volumes, difficult terrain and other factors may increase transportation costs to 20 USD/t CO2.The examples for Snohvit and Karstø discussed earlier show that transportation costs can besubstantial. Work to assess future transportation costs deserves more attention.


Table 3.14

Overview of likely CCS costs

Activity Cost (USD/t CO2) Uncertainties

CO2 capture 5 to 50 (current) Low end for pure streams that only need compression;

(including compression) 5 to 30 (future) high end for chemical absorption from gas-fired combined cycles

CO2 transportation 2 to 20 Depends on scale and distance

CO2 injection 2 to 50 Low end for Mt size aquifer storage;

high end for certain ECBM projects

CO2 revenues -55 to 0 No benefits for aquifers; highest benefits for certain EOR projects

Total -40 to 100

Note: Costs are expressed per tonne of CO2 avoided – see box on evaluating the cost of CCS in this chapter for conversion factors tocost per tonne of CO2 captured.

037-100 chapter 3 18/11/2004 07:20

Costs for injecting CO2 into depleted oil and gas reservoirs or aquifers are generally low. However,as the discussion of the Dutch ECBM project showed (Bergen et al., 2003), this is not always thecase. Cost depends both on the storage technology and on local conditions (e.g., if deviateddrilling is needed).

Part of the cost of CCS could be offset by revenues from enhanced fossil-fuel production. Thesebenefits could reach 55 USD/t CO2. Revenues from EOR in particular could be substantial, butthis is highly site specific and will not be the case for most CCS projects.

Total CCS costs can range from a 40 USD/t benefit in the most optimistic case to a 100 USD/tcost in cases of small-scale projects capturing CO2 from gas-fired power plants using existingtechnology (Table 3.14). This wide range shows that a case-by-case evaluation is needed for a propercost assessment. At this stage, for a vast majority of options, the total cost of CCS could be within50 to 100 USD/t CO2. By 2030, these costs should go down to 25-50 USD per tonne of CO2.


037-100 chapter 3 18/11/2004 07:20

037-100 chapter 3 18/11/2004 07:20

Chapter 4. BASIC RESULTS FROM THE MODEL ANALYSIS

H I G H L I G H T S

■ The Energy Technology Perspectives (ETP) model has been used to assess the potentialrole of CCS. ETP is a technology rich model that makes it possible to calculate the least-cost energy system for the period 2000-2050. The model results show that CCS compareswell with other technology options to reduce CO2 emissions.

■ In a scenario without new CO2 policies (which is based on the World Energy Outlook2004 Reference Scenario up to 2030), emissions increase one-and-a-half-fold from currentlevels to over 60 Gt CO2 by 2050. This increase is driven by strong economic growth andhigh coal growth rates.

■ A scenario with a 50 USD/t CO2 emission penalty (GLO50) results in a stabilization ofglobal emissions in the range of 23-28 Gt CO2/yr, which more than halves BASE emissionsin 2050. CO2 capture and storage increases to 18.4 Gt in 2050. This result should beconsidered as an upper limit for the CCS potential.

■ Using CCS achieves a 25% deeper cut in global emissions compared to the same GLO50scenario without CCS.

■ In the GLO50 scenario in 2020, 28% of total capture is from coal-fired processes. However,the share of coal increases to 65% by 2050, the remaining 35% being CO2 capture fromnatural gas, oil and biomass-fired processes, and to a lesser extent capture from cementkilns. The high coal share indicates the important synergy of CCS and coal.

■ CO2 capture in the electricity sector represents around 80% of total CO2 capture potentialin 2050. The remainder is evenly split between manufacturing and fuels production.

■ At a penalty level of USD 50/t CO2, 39% of all electricity production would be from plantsequipped with CO2 capture in 2050, including those that co-combust biomass.

■ Without CO2 policies, the average global efficiency of coal-fired power plants increasesfrom 32.1% to 42.7% in 2050, an increase of 10.6 percentage points, or 33%. Theefficiency of gas-fired plants increases from 36.0% to 57.4%, an increase of 21.4 percentagepoints or 59%. These efficiency gains are driven by technological progress and rising fuelprices, and they make CCS a viable option. However, due to the additional energy needsfor CCS, global average plant efficiency is in fact 3-6% lower (a 1 to 3 percentage pointreduction) in the GLO50 scenario with CCS.

■ Some 80% of all CO2 capture in the electricity sector is from IGCC type power plants. Upto 15 EJ of biomass is co-combusted in coal-fired IGCCs. According to the model, IGCCplants that co-generate electricity and synfuels pose an interesting option.

■ If CO2 penalties are introduced, ageing fossil-fuelled power plants with low efficiencyand without CO2 capture would either be closed down before the end of their technicallifespan or operate as peaking units. This indicates that the potential rate at which CCScan be introduced exceeds the rate at which capital stock is typically replaced. Regular

4. BASIC RESULTS FROM THE MODEL ANALYSIS 101

101-124 chapter 4 18/11/2004 07:24


capital stock turnover projections for the electricity sector, based on historical turnoverrates, do not apply if an ambitious CO2 policy is put in place.

■ In 2030, 80% of all CO2 capture takes place in OECD countries. By 2050, this share hasdeclined to 60%, if CO2 policies are introduced worldwide. This result is a function of theregional growth distribution of electricity demand, and global CO2 policy scenarios.

■ In 2030, half of the captured CO2 is used to enhance fossil-fuel production or stored indepleted oil and gas reservoirs, and half stored in aquifers. By 2050, aquifer storagedominates.

■ The additional emission reduction in GLO50, compared to the same CCS scenario withoutCCS, is equal to 40-45% of the quantity of CO2 captured. The fact that the emissionreduction is so much lower can be attributed to the additional energy use for CO2 captureand pressurization and the related emissions, and the increased coal share in the energymix.

■ The marginal emission reduction cost in the period 2030-2050 is halved, and the averageemission reduction cost declines by about a third (from about 45 to 30 USD/t CO2,) ifCCS is considered, compared to a scenario without CCS. This suggests that applying CCSwould result in a significant decrease in the policy incentives needed and it would alsosubstantially reduce the cost of CO2 policies.

This chapter starts with a brief description of the ETP model, followed by a discussion of the modelanalysis structure.

The ETP BASE scenario is presented, as this is the reference to which the other model scenarios arecompared. This is followed by a detailed discussion of CCS in the GLO50 Scenario. In this scenario,where the CO2 penalty is set at 50 USD/t CO2, global emissions are roughly stabilized at 23-28 Gt CO2 per year. The analysis shows that CCS can compete with other energy technology options,and that CCS technologies should form a key part of a CO2 emissions abatement scheme. CCS usein three key sectors is discussed in more detail: power generation, manufacturing and fuel processing.

The chapter ends with an analysis of CCS benefits in terms of emission reduction and cost of CO2

policies, based on a comparison of model runs with and without CCS.

The Energy Technology Perspectives (ETP) Model

CCS is a technology option for emissions reduction. The competition between CCS and other emissionreduction options is a complex issue. Its proper quantitative analysis requires a model that can dealwith technological change. A number of such models exist. The model used in this study is calledthe Energy Technology Perspectives (ETP) model. It belongs to the MARKAL family of bottom-upsystems engineering economic models (Fishbone and Abilock, 1981; Loulou et al., 2004). MARKALhas been developed over the past 30 years by the Energy Technology Systems Analysis Programme(ETSAP), one of the IEA Implementing Agreements (ETSAP, 2003).

Any model is a highly stylized representation of the world energy supply and demand, based on adataset that approximates the real world. Each model has its own unique characteristics that mayaffect the results and conclusions. A different model may produce different figures. Therefore the

101-124 chapter 4 18/11/2004 07:24

goal is to ‘model for insights’, not to ‘model for figures’, and each result should be interpreted withthe uncertainty of the results in mind. In this study special attention has therefore been given touncertainty analysis.

The ETP model is a micro-economic representation of part of the world economy (a so-called partialequilibrium model), divided into 15 regions. Only that part of the economy which is relevant toenergy is modelled. This so-called energy system is modelled as a set of interdependent technicalproduct flows and processes. Various technologies, characterized by their physical and economicproperties, can be used to generate certain product flows. A brief overview of the model technologydatabase is given in Annex 1.

The model process technology choice and process activity levels determine the physical and monetaryflows within the energy system. ETP is a linear programming model that minimizes the systems cost,given a certain demand for energy services and certain constraints, such as availability of naturalresources. Obviously, this is an abstract representation of the real world, where decisions are oftennot based on the same rigid cost minimization approach. Therefore, the primary goal of the ETPmodel is to identify optimal options and strategies and to assess the future role of energy technologies.It is not a tool for generating accurate energy projections.

Overview of the Model Analysis Structure

Figure 4.1 provides an overview of ETP model runs and their use for the analysis in Chapters 4 to7. A total of 35 model runs show a wide range of CCS potential and provide insights into the mainuncertainties that surround these modelling outcomes. The goal is to identify the factors which


ETP model analysis: caveats

The following ETP analysis discusses the potential for future use of CCS. It is not a predictionof what will happen, but rather an analysis based on the assumption of rational decision-making, a perfect market and perfect foresight. Risk and uncertainty are not accountedfor. The system is optimized based on cost considerations only. By definition, any CO2 emissionreduction option that would reduce systems cost would be part of the BASE scenario. Inreality, such a potential may exist, for example, through certain energy-efficiency measureson the demand side. Issues such as the uncertain permanence of CO2 storage, legal andregulatory barriers, and public acceptance of CCS, have not been considered in the model.

Apart from methodological caveats, there are also caveats of scope. Emission reductionoptions for greenhouse gases other than CO2 have not been accounted for. Land use, landuse change and forestry options (LULUCF) for CO2 emissions reductions are not consideredeither. If both of these emissions reductions were accounted for, the potential or need forCCS would decrease.

Finally, the analysis does not take into account the intra-regional distribution of emissionsources and potential storage sites. Consideration of such issues may reduce the potentialfor CCS. Given these limitations, model results for CCS use should be considered as optimisticpotentials, actual CCS use will be lower.

101-124 chapter 4 18/11/2004 07:24


are essential for the future application of CCS. The results map the potential range of CCS use,help identify key CCS technologies, and assess the impact on different world regions. CCS impactsare analysed from the perspective of the three shared goals of the IEA: energy security, acceptableenvironmental impacts and affordable energy.

The discussion is split into five main parts: CCS potential, CCS benefits, sensitivity analysis, scenarioanalysis and fuel market impacts. The analysis starts by looking at the ETP’s BASE scenario withoutCO2 policies. This scenario is discussed in order to allow comparison with other model analysisstudies. This is followed by the presentation of a scenario with a global penalty that gradually

Figure 4.1

The role of ETP model runs in the analysis

CHAPTER 4

Includes a discussion of impacts of individual parameter uncertainty

on CCS use

Includes a discussion of impactsof combined parameter uncertainty on CCS use

Includes impacts on fuel markets of having CCS

CHAPTER 5

Scenario AnalysisCHAPTER 6

Energy Market Impacts

The justification for CCS. Includes:• The results of GLO50 with & without CCS• A discussion of CCS versus other strategies (renewables, nuclear, energy efficiency)

CHAPTER 7

BASE

GLO50with CCS

EFTEP+ + + + +

EFTEP+ – – – –

EFTEP– + + – –

EFTEP– + – + +

GLO10w&w/o

CCS

GLO25w&w/o

CCS

GLO50w&w/o

CCS

GLO100w&w/o

CCS

21 model runs

GLO50w/o CCS

GLO50 w & w/o CCS + NUC (4 runs)

CCS Potentials AnalysisCHAPTER 4

GLO50 w CCS(reference)

CCS Benefit Analysis

Sensitivity Analysis

101-124 chapter 4 18/11/2004 07:24

increases to USD 50/t CO2, named the GLO50 scenario. A penalty is an abstract way of representingregulatory and financial policy instruments. The 50 USD/t CO2 penalty level was chosen for moredetailed discussion because it roughly represents emission stabilization in the period 2000-2050,or a halving of emissions by 2050, compared to the BASE scenario. Furthermore, it is clear fromprevious analysis that CCS is a costly option compared to other greenhouse gas emission reductionoptions.

The GLO50 discussion focuses on CCS use by sector and on storage. This is followed by an analysisof CCS benefits. Benefits are considered in terms of environmental benefits (the additional emissionsreduction due to CCS) and financial benefits (reduced cost to achieve certain emission reductiontargets). For this purpose, a GLO50 case without CCS is compared to the one with CCS, and fourmodel runs with and without nuclear and CCS are also compared.

In 21 sensitivity model runs, a range of potentially important parameters was varied for the GLO50scenario in order to assess which of these are crucial for the future role of CCS. Next, four scenarioswere defined along the lines of these key parameters. These scenarios are characterized by theacronym EFTEP: Economy (E), Fuel demand and price (F), Technological progress (T), Environment(E) and regional Policy scope (P). A plus (+) means that the parameter values of the scenariodimension result in high CCS potential, while a minus (-) indicates low CCS potential. These scenariosshow how the interactions of positive and negative factors affect the potential for CCS use. Theresults can be used to identify which scenario dimensions are more important than others.

The fuel market consequences of CCS are analysed through a structured set of model runs withpenalties of 10, 25, 50 and 100 USD/t CO2, with and without CCS. Fuel use and prices are mappedfor each of these scenarios. The differences provide insights into supply security consequences.

The BASE Scenario

The BASE scenario is the scenario against which all other scenarios in this book are evaluated. Thescenario is based on the development in the World Energy Outlook 2004 Reference Scenario upto 2030 (IEA, 2004a). It includes, as the Reference Scenario, energy and climate policies enactedbefore mid-2004, as well as further policies beyond that. Refer to Chapter 2 for more discussionof the Reference Scenario.

For the period 2030 to 2050, the BASE scenario is a result of extrapolated demand projectionsand technology and fuel choices driven by the model algorithm and the technology assumptions.No additional polices are implemented beyond those included in the Reference Scenario. Giventhe fact that CCS will incur additional costs, there is thus no role for CCS in this scenario.

Primary energy demand in the BASE scenario more than doubles in 50 years (Figure 4.2). Thegrowth is mainly accounted for by coal, and to a lesser extent by natural gas and oil. In theabsence of CO2 policies, this scenario therefore implies a continued reliance on fossil fuels. Thehigh growth-rate for coal can be explained by slower growth in coal prices compared to oil andgas prices, and introduction of new coal conversion technologies. CO2 emissions in this scenarioincrease substantially because of the high growth-rate for coal. The important role of coal in thisscenario implies considerable potential for CCS if proper regulations or financial incentives areput in place.

In the BASE scenario electricity production almost triples in the period 2000-2050. The bulk ofelectricity production growth is accounted for by fossil fuels. Production of electricity from coal more


101-124 chapter 4 18/11/2004 07:24


than doubles, while electricity production from natural gas quadruples. Nuclear electricity productionincreases, but at a low rate. Production of electricity from renewables quadruples over 50 years.The share of renewables in the electricity mix increases from 19% in 2000 to 28% by 2050. Thegrowth in renewables is accounted for by hydro, biomass/waste, geothermal and wind.

The GLO50 Scenario

This section discusses the CO2 policy scenario entitled GLO50. The name GLO50 refers to the factthat this scenario includes a global CO2 penalty at the level of 50 USD/t CO2. The goal is to showthe competitiveness of CCS in the three main application fields: electricity production, manufacturingindustry and fuels supply. The discussion will allow the reader to understand better the ETP sensitivityanalysis results outlined in the next chapter. Four specific scenarios are then discussed in Chapter6, based on the analysis of the GLO50 results and the sensitivity analysis. These scenarios outlinethe future potential for CCS. It is worth noting at this stage that the GLO50 scenario should notbe considered a ‘best guess’ scenario since a wide range of uncertainties can affect the results.Given the high BASE emissions, optimistic assumptions for CCS and conservative assumptionsfor competing options in this scenario, the CCS results in GLO50 should be considered as anupper limit for the CCS potential.

In a bottom-up model such as ETP, various types of policies can be simulated. For the analysis inthis section, CO2 policies are simulated by imposing CO2 emission penalties. These CO2 emission

Figure 4.2

Primary energy demand projections in the BASE scenario

Key point: If no CO2 policies were introduced, coal would account for the bulk of the increase in primary energy supply

in the BASE Scenario over the next 50 years

1000

900

800

700

600

500

400

300

200

100

0EJ/l

yr

2050

2045

2040

2035

2030

2025

2020

2015

2010

2005

2000

Renewables

Nuclear

Gas

Coal

Oil

101-124 chapter 4 18/11/2004 07:24

penalties are invariably the costs incurred to deploy the relevant technologies (e.g., because ofregulations), but they can also be interpreted as the level of a tax on CO2 emissions or the priceof a tradable emissions permit on the market. According to standard economic reasoning, firmsconfronted with such ‘prices’ for GHG emissions will deploy all technologies that cost less than these‘prices’. Penalties are expressed in USD per tonne of CO2 equivalent. They apply to all GHG emissionsfrom the energy system.

Figure 4.3 shows the CO2 penalties and the date at which they are introduced into the GLO50scenario for each region. It is assumed that policies in developing countries are introduced at a laterstage than in industrialized countries. The model input data specify that the penalty in industrializedcountries starts in 2005, reaches the level of 50 USD/t CO2 by 2015, and stabilizes thereafter. Indeveloping countries, the policy is introduced in 2020 with the penalty reaching its maximumlevel by 2030.

While 50 USD/t CO2 may seem a high burden for developing countries, it is not impossible thatsuch penalty levels are applied in the long term, given the environmental concerns and the economicdevelopment potential. In the model scenarios, by 2050 per capita GDP in all regions exceptAfrica is close to or higher than the per capita GDP in OECD Europe in 2000 (see Annex 3).


Figure 4.3

GHG penalties in the GLO50 scenario

Key point: In the model analysis, CO2 penalties are introduced by industrialized countries from 2005, reaching the level of 50 USD/t CO2

by 2015 and stabilizing thereafter. In developing countries, the policy is adopted 15 years later

USD

/t C

O2

60

20

10

0

40

30

50

2010

2000

2020

2030

2040

2050

Developing countries

OECD and Transition Economies

101-124 chapter 4 18/11/2004 07:24


Figure 4.4 shows the CO2 emissions in the BASE scenario and the GLO50 reference scenario. Notethat in the BASE scenario, the growth of CO2 emissions between 2000 and 2050 is 1.8% peryear, the same as for the period 1971-2000. The GLO50 scenario represents roughly a stabilizationof global emissions at a level of 23-28 Gt CO2/yr, which halves BASE scenario emissions in2050. This is a significant emissions reduction. A more detailed discussion in Chapter 5 will showthat this scenario could be consistent with a stabilization of the atmospheric concentration of CO2

at 550 parts per million (ppm).

Figure 4.5 shows the primary energy mix in the GLO50 scenario. Total primary energy use is about850 EJ in 2050. This is some 8% lower than in the BASE scenario. This decline is the net result offuel switching (which increases energy efficiency), demand-side energy efficiency measures, andincreased energy use for CO2 emission mitigation measures such as CCS.

In the GLO50 scenario, coal use is stable up to 2030, but shows strong growth beyond this date.Both coal and oil use in 2050 are significantly lower than in the BASE scenario. Gas use is virtuallythe same as in the BASE scenario. The use of renewables increases by 80%. Biomass use doublesand the use of wind more than doubles by 2050, compared to BASE. Nuclear shows no growth,but this is largely explained by the constraints applied to this energy source, in line with the WorldEnergy Outlook (IEA, 2004a).

Figure 4.4

Global CO2 emissions, BASE and GLO50 scenarios

Key point: Emissions can be stabilized at a penalty of 50 USD/t CO2

Gt

CO

2/y

r

70

20

10

0

40

30

60

50

1980

1970

2000

1990

2010

2020

2030

2040

2050

GLO50

Statistics

Base

Note: Historical statistical data exclude CO2 emissions from limestone calcination in cement production and other industrial processemissions, which explains the gap in 2000.

Source: IEA, 2002b.

101-124 chapter 4 18/11/2004 07:24

The results for CCS in the GLO50 scenario represent the upper limit of the potential role of CCS.This scenario includes speculative CCS technologies, but conservative estimates for competingemissions reduction options such as renewables and nuclear. The impact of different assumptionson CCS use is discussed in more detail in Chapter 5.

Figure 4.6 shows CO2 capture in the GLO50 scenario by process area. Capture technology beginsto be applied around 2015, increases to over 8 Gt by 2030 and to more than 18 Gt by 2050.These quantities should be considered as potentials, not as projections. The amount that is capturedand stored can be compared to the 34 Gt of CO2 emission reductions in this scenario by 2050,compared to the BASE scenario (Figure 4.4). The figures suggest that CCS represents a significantshare of total emissions reduction. Note that the growth in CCS capacity between 2020 and 2030is very rapid and may be unrealistic in terms of yearly expansion of the industry. No growth constraintshave been applied to account for capacity expansion limitations such as regulatory proceduresand slow growth of public acceptance.

Capture from power plants increases at a faster rate than capture in the manufacturing industry.The share of capture from power plants increases from 53% in 2020 to 80% of total CO2

capture in 2050. This includes capture from industrial CHP installations and from plants whichco-generate electricity and synfuels.


Figure 4.5

Primary energy mix in the GLO50 scenario

Key point: Compared to the BASE scenario, a penalty of 50 USD/t CO2 results insignificantly higher renewable energy use and a reduction

in the use of coal

1000

800

600

300

900

700

500

100

400

200

0EJ/y

r

2000

2030

BASE 20

30

GLO50 20

50

BASE 20

50

GLO50

Nuclear

Renewables

Gas

Coal

OilOil

101-124 chapter 4 18/11/2004 07:24


Out of the total quantity captured that is shown in Figure 4.6, 28% is from coal-fired processesin 2020. However, the share of coal increases to 65% by 2050. The remaining 35% is capturefrom natural gas, oil and biomass-fuelled processes, and to a lesser extent capture from cementkilns. This distribution indicates the importance of CCS for the future of coal in a CO2-constrainedscenario.

Figure 4.7 shows a subdivision of CO2 capture by technology. The results suggest that IGCCtechnology will play a key role in CCS. This is closely related to the dominance of coal in the fuelmix. In this analysis, IGCC includes plants which co-generate electricity and synfuels. Chemicallooping plays a secondary role compared to IGCC, while steam cycles with flue gas CO2 captureare not selected. This result depends on the technology data assumptions. The impact of lessoptimistic assumptions for IGCC is discussed in Chapters 5 and 6.

Up to 2025, capture is concentrated in industrialized countries. Beyond 2025, capture indeveloping countries grows at a high rate. By 2050, 46% of total capture activity is undertakenin developing countries. This pattern can be explained by the assumption that there is a delayedintroduction of CO2 policies in developing countries and by the fact that growth in emissions overthe next 50 years is concentrated in such countries. The high share of capture in developing countriesin this scenario suggests that if CCS is not applied in developing countries, the total quantitycaptured worldwide will be much lower. This indicates that international co-operation regardingCO2 emission mitigation is needed for the widespread uptake of CCS technology.

Figure 4.6

Global CO2 capture by process area, GLO50 scenario

Key point: Capture from power plants represents four-fifths of the cost-effective capture potential

20

10

8

6

14

12

16

18

4

2

0Gt

CO

2/y

r

2050

2045

2040

2035

2030

2025

2020

2015

2010

2005

2000

Fuelprocessing

Manufacturingindustry

Electricity production

Note: The category electricity production includes capture from plants which co-generate electricity and synfuels as well as industrialCHP plants.

101-124 chapter 4 18/11/2004 07:24


Figure 4.7

Total CO2 capture split by technology, GLO50 scenario

Key point: IGCC with physical CO2 absorption is the dominanttechnology deployed to capture CO2 from power generation processes

20

18

16

14

12

10

8

4

2

6

0Gt

CO

2/y

r

2050

2045

2040

2035

2030

2025

2020

2015

2010

2005

2000

Other

Chemical looping

NGCC

IGCC

CCS and the WEO 2004 Alternative Policy Scenario

The World Energy Outlook 2004 Alternative Policy Scenario (APS) analyses the impact offaster deployment of many different types of end-uses and supply technologies. They rangefrom hybrid vehicles to power generation fuel cells, from water solar heaters to distributedgeneration. The impact of the introduction of CCS and other breakthrough technologies isnot included in the Alternative Policy Scenario. However, WEO 2004 presents an examplewhere CCS is included as an add-on to the APS development.

In the Alternative Policy Scenario, about 136 GW of new coal-fired power-generation capacityand 38 GW of new centralized gas-fired capacity are expected to be built in OECD countriesbetween 2015 and 2030. New capacity additions in the transition economies and thedeveloping countries are larger. The WEO-CCS example assumes that all new capacity inOECD countries built after 2015 is equipped with CO2 capture technology and that thiscapacity is matched by a similar amount in non-OECD countries. By 2030, the reductionin CO2 emissions would be between 1.5 and 2 gigatonnes, depending on the utilizationrate of the power plants and the energy consumed in capturing and pressurizing the CO2.Taking the average of 1.75 Gt, the total emission reduction in the APS would increase from16% to 21% compared with the Reference Scenario.1 CCS would in this case cover 5% oftotal world generation capacity in 2030, compared to 36% for renewables-based generation.

1. This analysis does not take into account the cost-effectiveness of CCS compared with other options for reducing emissions, such asenergy efficiency and renewables.

101-124 chapter 4 18/11/2004 07:24


CO2 capture in the electricity sectorIn the GLO50 scenario with CCS, electricity production capacity almost triples from 3.5 TW in2000 to 10.3 TW by 2050. This includes industrial CHP plants and plants that co-produce electricityand synfuels. Power plants with CO2 capture represent 22% of total installed capacity by 2050.This is almost half the total installed capacity of fossil-fueled power plants.

The electricity production mix in the GLO50 scenario is shown in Figure 4.8. Total production amountsto 153 EJ in 2050, compared to 148 EJ in the BASE scenario. Electricity production almost triplesin the period 2000-2050. The small increase in GLO50 compared to BASE in 2050 can be explainedby a higher share of electricity in the final energy demand. Additional electricity use for CCS is notaccounted for explicitly; it will show up as a reduction in the power plant efficiency and, therefore,as increased fuel use in the electricity sector.

The production share of fossil-fueled power plants with CCS increases to 39% by 2050. Thisincludes coal-fired power plants where biomass is co-combusted. If the share is corrected for biomassco-combustion, fossil fuels with CO2 capture represent 37% of total electricity production. The shareof CCS in electricity production is much higher than the share of electricity production capacity.This can be explained by the high load factor for these facilities with CO2 capture, and adeclining load factor for plants without capture. The reason for this change in load factors isthat plants without CO2 capture are operated as middle-load and peaking plants, while plantswith CCS are operated as base-load plants. This, in turn, can be explained by the comparativelyhigh marginal cost of power production from fossil fuels without CO2 capture, if a CO2 penalty isintroduced. The share of fossil fuels in electricity production declines from 64% in 2000 to 54%in 2050. This is compensated by an increased share of renewables in the electricity mix, whichincrease from 19% in 2000 to 40% by 2050. The main increase of renewables comes from wind,biomass and hydro.

The potential for CCS in this example is significantly lower than in the GLO50 scenario. Inthe GLO50 scenario more than 900 GW of power generation capacity is equipped withCCS technology by 2030, of which 580 GW is coal-fired resulting in 8.3 Gt CO2 beingcaptured. There are several reasons for the difference between the WEO Alternative PolicyScenario example and the GLO50 scenario: the most important is that in the APS a largeshare of new OECD generating capacity is based on decentralized gas-fired generationand there is thus very modest growth in new centralized coal generation after 2015. Sinceonly centralized plants are considered to be equipped with CCS technology in this example,this significantly reduces the potential. In the WEO Reference Scenario the growth incentralized gas and coal-based power generation is much higher. If, instead, CCS had beenconsidered as an add-on to this scenario, it would have shown a potential for 4.7 Gt CO2

capture by 2030.

Moreover, in the WEO analysis CCS is only considered for new centralized power plantsand is, therefore, excluded from industry, fuel-processing plants and electricity and synfuelcogeneration plants. In 2030, 29% of the total amount of CO2 that is captured in theGLO50 scenario is outside power generation. This share declines to 20% by 2050. Withinthe electricity sector, in the GLO50 scenario plants that co-generate electricity and synfuelsaccount for 2 Gt capture in 2030, rising to 10 Gt CO2 capture in 2050. This option doublesthe capture potential in the electricity sector, and it was not considered in the WEO analysis.

101-124 chapter 4 18/11/2004 07:24


Figure 4.8

The electricity production mix, GLO50 scenario

Key point: Electricity production from power plants equippedwith CCS increases to over a third of total production by 2050

160

140

120

100

80

60

40

20

0EJ e

lect

rici

ty/y

r

2050

2045

2040

2035

2030

2025

2020

2015

2010

2005

2000

Geothermal

Wind

Biomass/ waste

Hydro

Nuclear

Fossil fuels with CCS

Fossil fuels w/o CCS

The future of decentralized power generation

Decentralized power generation has received considerable attention. During the past decade,most attention was focused on micro turbines and Stirling engines and renewables. In recentyears, attention has switched to fuel cells. CHP is the largest existing decentralized market.Large-scale CHP systems (based on gas turbines and boilers) represent 96% of the CHPmarket worldwide (WADE, 2003). Decentralized power supply systems based on renewableenergy have been introduced in developing countries where a transmission and distributionsystem is lacking. Sparsely populated regions of industrialized countries, high grid connectioncosts and the availability of renewable resources can also result in a switch to decentralizedgeneration (Swisher, 2002). However these conditions apply to niche markets with differentcharacteristics from densely populated urban areas, where security of supply and higherenergy efficiency would be the main advantages of CHP use.

While CHP systems have reached maturity on a 1 MW+ scale, they are not yet widely appliedon a smaller scale although there is potential in the residential and commercial sectors.Their main appeal would be savings in transmission costs and losses. However, the 40-50% electric efficiency of a gas-fired fuel cell is much lower than the 60% electric efficiencyof new centralized gas-fired power plants. In CHP mode there are heat benefits of fuel cellsystems but such systems require applications with a continuous heat demand, not onlyseasonal space heating or cooling.

101-124 chapter 4 18/11/2004 07:24


However, fuel consumption for electricity production remains stable up to 2025. This can be explainedby a significant efficiency gain for fossil-fueled power plants during the period 2000-2025. Moreover,in the IEA accounting the primary energy equivalents of wind, geothermal and hydro electricityare equal to the amount of electricity produced. This is equivalent to a conversion efficiency of100%. Therefore, switching from fossil fuels to renewables in the period 2000-2050 adds topower sector energy efficiency gains. The combined effect of increased production and efficiencygains is a stabilization of fuel use up to 2025.

From 2025 to 2050, fuel use doubles to reach 330 EJ. The reason for this is that electricity andtransportation fuel cogeneration plants are introduced. These plants use more fuel per kWh ofelectricity produced, which results in increased fuel use. Moreover, these cogeneration plants usecoal as a fuel. The increased share of coal in the electricity mix also results in a drop in averageefficiency.

Figure 4.9 shows efficiency trends for coal and gas-fired power plants. CHP plants are excluded fromthis analysis. For plants that co-generate electricity and transportation fuels, a correction has beenapplied, based on fuel use for stand-alone synthetic transportation fuel production. The efficienciesin 2000 are gross efficiencies; net efficiencies are 1-2 percentage points lower.2 All new power plantsin the ETP model are characterized on a net basis, meaning that in 2050 all efficiency figures areon a net basis.

The model suggests important efficiency gains for coal and gas-fired power plants in the BASEscenario. For coal, global average efficiency increases from 32.1% in 2000 to 43.2% by 2050.For gas, average efficiency increases from 36.1% in 2000 to 53.8% by 2050. These efficiency

Fuel cells for CHP have long-term potential (Pehnt, 2004). It can be argued that decentralizedsystems increase supply security, but this will depend on the reliability of the technologyapplied. Current decentralized systems (usually diesel engines) are often operated as back-up systems, so they do indeed increase reliability. However, it is debatable whether adecentralized system without a grid connection would increase security of supply. Anotherdisadvantage is that CCS would not be feasible for small-scale fuel cell systems. If zeroemissions in power generation are the aim, a hydrogen supply system would be needed,similar to existing natural gas pipeline systems.

In its Reference Scenario, the IEA World Energy Outlook projects 98 GW of fuel cell capacityby 2030. This represents about 1.3% of global electricity capacity (IEA, 2004a). In theAlternative Policy Scenario, global electricity generation from fuel cells using hydrogenfrom reformed natural gas is 530 TWh, twice as high as in the Reference Scenario in 2030.This projection suggests that hydrogen demand will be limited. Moreover, in regions wherehydrogen can be supplied by pipeline, electricity transmission must be a viable alternative.Distributed generation based on hydrogen fuel cells is, therefore, considered a topic ofsecondary importance for CO2 emission reduction. A rapid expansion of distributed renewableelectricity generation, especially in developing countries, could limit the need for newcentralized capacity and, therefore, reduce the need for CCS.

2. Net efficiencies cannot be tracked from the IEA statistics because own electricity use by power plants is only reported as an aggregatefor all fuels.

101-124 chapter 4 18/11/2004 07:24

gains are the result of the replacement of old low-efficiency power plants, technologicalimprovements, and increasing fuel prices that result in a switch to higher-efficiency power plantspurely for economic reasons.

For gas, and to a lesser extent for coal, gains in the GLO50 scenario are smaller than in the BASEscenario because of additional power-plant electricity use for CO2 capture and pressurization. Theefficiency loss is to some extent compensated for by choosing more efficient power plant technologies.The efficiency level shown in Figure 4.9 is an average for power plants with and without CO2 capture.This average efficiency is in fact 3-6% lower (a 1 to 3 percentage point reduction) in a scenariowith CCS.

Note: Coal includes hard coal and lignite. Corrected for the co-production of synfuels. Efficiencies in 2000 are gross efficiencies, netefficiencies are 1-2 percentage points lower. Efficiencies in later decades are net efficiencies.

Figure 4.10 shows electricity production from power plants fitted with CO2 capture. Total productionamounts to 27 EJ in 2030 and 64 EJ by 2050. Coal-fired power plants represent 60% of totalelectricity production capacity fitted with CCS in 2030. This percentage increases to 69% by2050 (25% of total power production). Gas-fired power plants represent 28% of all powerplants with CCS in 2030, declining to 23% by 2050 (8.4% of total power production). Theremainder is dedicated biomass-fired power plants fitted with CO2 capture. Significant amounts ofbiomass are also co-combusted in coal-fired power plants, meaning that the share of coal with CO2

capture is in fact somewhat lower and the share of biomass somewhat higher.

From 2025 onwards, IGCC plants producing both electricity and synfuels show strong growth.Three-quarters of the synfuel produced is hydrogen, while the remainder is DME. Hydrogen isused in equal parts by the transportation sector and by industry as a substitute for natural gas.


Figure 4.9

Efficiency trends for coal and gas-fired power plants

Key point: Future power plant efficiency increases by a third for coal and by two-thirds for gas

Effic

ien

cy (

LHV

) (%

)

2050

2045

2040

2035

2030

2025

2020

2015

2010

2005

2000

0

10

20

30

40

50

60

70

Coal, GLO50

Coal, BASE

Gas, GLO50

Gas, BASE

101-124 chapter 4 18/11/2004 07:24


Note that hydrogen and DME are the only two synfuels for which co-production with electricity isconsidered. In principle, other synfuels such as methanol, low-sulphur diesel and naphtha couldalso be produced (Espinoza et al., 1999; Steynberg and Nel, 2004). These fuels could be used inthe existing transportation infrastructure. However, combustion of hydrocarbon synfuels results inCO2 emissions. Overall, the CO2 benefits will therefore be smaller if a carbon-containing synfuel isproduced than if hydrogen is produced. So far, South Africa is the only country that produces synfuelsfrom coal on a large scale, although China has announced its intention to produce 60 Mt of synfuelsfrom coal by 2030, based on Sasol technology. Such plants would produce fuels and electricity inan 8:1 ratio in energy terms and achieve 46% overall energy efficiency (Steynberg and Nel, 2004).

The plants considered in this analysis would produce synfuels and electricity in a 2:1 ratio (forDME production) or in a 2:3 ratio (for hydrogen), and achieve an overall energy efficiency of 50-53%. Results could be affected by different technology assumptions. In a sensitivity analysis theimpact of synfuel cogeneration technology has been analysed in more detail (see Chapter 5).

CO2 capture in the manufacturing industryFigure 4.11 shows CO2 capture in the manufacturing industry. Capture from manufacturing is anorder of magnitude smaller than capture from electricity production plants. It can be split intothree parts: ammonia production, cement kilns, and iron and steel production (blast furnacesand DRI production). These three sources are of similar importance. While capture from ammoniaand DRI plants is based on established technology, capture from cement kilns and blast furnacesis a new concept that may require major process adjustments. The future role of these sources is

Figure 4.10

Electricity production from power plants fitted with CCS,by technology and fuel, GLO50 scenario

Key point: IGCC plants used for electricity and synfuelcogeneration dominate total power plant capacity fitted with CCS

70

60

50

40

30

20

10

0EJ e

lect

rici

ty/y

r

2050

2045

2040

2035

2030

2025

2020

2015

2010

2005

2000

Biomass blackliquor IGCC

Gas combined cycle

Coal other

Coal IGCC synfuel cogeneration

Coal IGCC

101-124 chapter 4 18/11/2004 07:24

therefore less certain than capture from ammonia plants. In the GLO50 scenario, CCS from ammoniaplants starts in 2010. This represents an early application of the technology with low capture cost.

While the CCS potential in the manufacturing industry is initially significant, in later decades it isrestricted by the limited growth of BASE scenario emissions. The growth rate of emissions in theelectricity sector is much higher. One reason for this is the assumed global trend towardsdematerialization of economic growth.

Note that in this study CHP plants fitted with CO2 capture are allocated to the electricity sector.The bulk of these plants would actually provide heat to industrial firms, and may be owned bysuch firms. Also, the results suggest some hydrogen delivery to industry for stationary use. Thishydrogen is produced from fossil fuels with CO2 capture. CO2 captured from the production of thishydrogen is allocated to the fuel-processing industry or the electricity sector. Given such linkages,one could argue that the use of CO2 capture in industry is in fact higher than that indicated inFigure 4.11.

CO2 capture in the fuels supplyFigure 4.12 shows CO2 capture potential from fuel production processes. The total quantity capturedis an order of magnitude smaller than that captured from the electricity sector. Roughly 40% iscaptured from hydrogen production processes that use natural gas, and another 40% is capturedin the Fischer-Tropsch synthesis of transportation fuels that use biomass and coal. The remainderis captured from refineries, particularly coking units for heavy residues and tar-sand processingplants.


Figure 4.11

CO2 capture in the manufacturing industry, GLO50 scenario

Key point: Capture from industry offers early opportunities, but has limited long-term potential

1.6

1.4

1.2

1

0.8

0.6

0.4

0.2

0Gt

CO

2/y

r

2050

2045

2040

2035

2030

2025

2020

2015

2010

2005

2000

DRI production

Blast furnaces

Cement kilns

Ammonia production

101-124 chapter 4 18/11/2004 07:24


As mentioned earlier, CO2 capture in cogeneration of electricity and synfuels is allocated to theelectricity sector. Capture from the cogeneration of electricity and synfuels is of much moreimportance than capture from dedicated synfuel production units. The capture from cogenerationplants amounts to 10 Gt CO2 by 2050, two-thirds of which represents capture from electricityand hydrogen cogeneration plants. The capture from these cogeneration plants represents 54%of total CO2 capture by 2050.

CO2 storageFigure 4.13 shows the results for CO2 storage under the GLO50 scenario. Storage is roughlyevenly divided between aquifers and depleted oil and gas fields, including enhanced oil andgas recovery operations (EOR and EGR). This is a result of the global distribution of potentialstorage sites and emission sources. Total cumulative storage over the period 2000-2050 amountsto 387 Gt, a small share of the total global storage potential, or roughly half the amount that canbe stored worldwide in depleted oil and gas reservoirs.

In a least-cost optimization model such as ETP, one might expect that CO2 use for enhanced fossilfuel production is chosen first. However, only 3% of the current world oil production is based onEOR. The remaining 97% is based on primary and secondary production technologies. The growthof EOR in general limits the growth of CO2 EOR. In fact, CO2 EOR has been applied on a limitedscale for the past 25 years, and opportunities are likely to increase gradually over the next 15 yearsas production in certain basins such as the North Sea and the Gulf of Mexico matures. A similar

Figure 4.12

CO2 capture in the fuels supply sector, GLO50 scenario

Key point: There is considerable potential for capturing CO2

from hydrogen production, fuel refining and FT synthesis processes

2.5

2

1.5

1

0.5

0Gt

CO

2/y

r

2050

2045

2040

2035

2030

2025

2020

2015

2010

2005

2000

Refineries

FT synthesisfrom coal andbiomass

Hydrogen productionfrom natural gas

FT = Fischer-Tropsch

101-124 chapter 4 18/11/2004 07:24

explanation can be given for the EGR model results. Also, CO2-EOR competes in practice withother EOR options. Many oil and gas fields are in remote regions, far from sources where CO2

could be captured. In such cases, the effort to bring CO2 to the site must be compared to the costof alternative EOR technologies. The model results regarding CO2 use for EOR are subject to significantuncertainties. A proper assessment of the potential would require detailed field-by-field data,which is beyond the scope of the ETP model analysis.

Wherever there is an opportunity to generate revenues using CO2, while achieving long-termstorage, such opportunities should be used. The results suggest that such opportunities arenot critical for the feasibility of CCS, however. This enhances the robustness of the results, assuch fossil fuel revenues constitute a source of uncertainty. The timing for the introduction ofEOR and EGR assumed in this analysis should be considered merely indicative, with an accuracyof ±10 years. With EGR, it is important to bear in mind that this is a speculative technology, andthat benefits will in most cases be small, compared to benefits for EOR. Therefore, storage in depletedgas fields and EGR are considered as one single category in Figure 4.13.

CCS Compared to other Emission Reduction Options

This section discusses the environmental and financial benefits of a CCS options, compared to otherCO2 emission reduction options in the energy sector. The analysis is split into two parts. First, theGLO50 scenario with and without CCS are compared. This builds on the analysis in the previoussection. Next, four model runs, with and without CCS and nuclear, are compared. This analysis


Figure 4.13

CO2 storage in the GLO50 scenario

Key point: Half of CO2 is stored in aquifers and the rest in oil and gas reservoirs

20

18

16

12

14

10

8

6

4

2

0Gt

CO

2/y

r

2050

2045

2040

2035

2030

2025

2020

2015

2010

2005

2000

Aquifers

EGR + depletedgas fields

EOR + depleted oil fields

EOR = enhanced oil recovery, EGR = enhanced gas recovery.

101-124 chapter 4 18/11/2004 07:24


provides insights regarding the financial benefits of having CCS, compared to other emissionmitigation options.

In order to assess the benefits of CCS, the GLO50 scenario and the same scenario without CCStechnologies (GLO50noCCS) are compared in this section. Comparing results at the same policyincentive level with and without CCS illustrate the benefits of CCS and its impacts on the energysystem.

Firstly, CO2 mitigation benefits are discussed. Figure 4.14 shows CO2 emissions in a GLO50 casewith and without CCS. The difference amounts to 4.6 Gt in 2030 and 7.9 Gt by 2050. WithoutCCS, CO2 emission reduction declines by about 20%, with emissions 28% higher compared tothe same GLO50 scenario with CCS.

Secondly, the economic benefits of CCS are discussed. In Figure 4.15, the cumulative emissionreduction for the period 2000-2050 is shown as a function of the CO2 penalty. With ambitiouspolicy targets, allowing for CCS cuts by half the penalty needed to reach a certain cumulativeemission reduction. When CCS is not considered, other emission reduction options can be appliedto reach the same targets, but the cost will increase. For example, the undiscounted cumulativesystems cost to reach the GLO50 scenario cumulative emission reduction without CCS increases by11 trillion USD, or 63%. This result does, however, depend critically on the technology learningassumptions for renewables and the ambitious policy target, and should be considered a high endestimate.

Figure 4.15 also shows the cumulative CO2 capture in the period 2000-2050. The quantity capturedequals some 43-58% of the total cumulative emission abatement. The area between the curveswith and without CCS is smaller and indicates that the actual emission reduction of CCS is only

Figure 4.14

CO2 emissions with and without CCS

Key point: Without CCS, total emission reduction potential declines by one-fifth

Gt

CO

2/y

r

70

50

30

40

20

10

0

60

GLO50

GLO50noCCS

Base

2000

2010

2020

2030

2040

2050

101-124 chapter 4 18/11/2004 07:24

40-45% of the quantity captured. This is due to the additional energy used for CO2 capture andpressurization and related emissions, and certain leakage effects (more coal use and less use ofother fuels with low associated CO2 emissions).

The cumulative capture increases with the penalty level. This shows that there is indeed no singlecost figure for CCS, as discussed in Chapter 3. Instead, a cost range exists. The shape of the curveindicates that the additional cumulative capture decreases for each USD increase of the penalty.Most of the CCS potential is below 50 USD/t CO2. For CCS, the additional capture in the case ofhigher penalties is limited. The impact of the penalty level on CCS use is studied in more detail inChapter 6.

The comparison of the GLO50 model results with and without CCS also provides insights into theimpact of CCS use on fuel markets. Chapter 7 studies this analysis in more detail. Without CCS,coal use declines over the next 50 years. With CCS available, coal use doubles. This increase iscompensated by a reduced growth of renewables and by reduced energy efficiency gains, comparedto a scenario without CCS. The results suggest that the fuel market consequences of CCS could besubstantial on a global scale.

Next, the global annual emissions are fixed as in the GLO50 scenario, and the set of options availableto reduce the emissions is varied. Four combinations with and without nuclear and/or CCS areanalysed: no CCS and nuclear (no NUC+CCS), CCS, NUC and CCS+NUC. In this approach, the CCScase is almost equivalent to the GLO50 case. Small differences can occur because the emissionconstraint that is imposed in the model in the CCS run is applied to the world as a whole, while


Figure 4.15

Cumulative emission abatement for 2000-2050 as a function of the penalty level

Key point: Up to 2050, the cumulative reduction in CO2

emissions is one-fifth lower if CCS is not considered than it would be if CCS was applied. This shows the environmental benefits of CCS

Gt

CO

2

1200

1000

600

800

400

200

0250 50 75 100

Total w/o CCS

CO2 capture

Total with CCS

Penalty (USD/t CO2)

101-124 chapter 4 18/11/2004 07:24


the penalty in the period 2005-2030 in the GLO50 scenario differs for industrialized countriesand developing countries. In the CCS run, the model chooses cheaper emission reduction optionsin developing countries in the period 2005-2030 instead of applying costly options in industrializedcountries that are selected in the GLO50 scenario. From 2030 onwards, however, GLO50 and theCCS case are virtually identical.

The NUC case allows for unlimited nuclear growth in OECD countries, and 10% annual growthpotential in developing countries, while nuclear growth potentials in all runs without NUC are halvedin developing countries, and a maximum nuclear use is defined for OECD countries, in line withthe WEO Reference Scenario (IEA, 2004). The actual nuclear investment depends on cost-effectiveness,compared to other zero emission strategies.

The resulting marginal emission reduction costs are shown in Figure 4.16. These show that indeedthe CCS case is almost equivalent to the GLO50 scenario, with the marginal cost from 2030 onwardsat 50 USD/t CO2. The average emission reduction cost is shown in Figure 4.17. As could havebeen expected, the lowest cost occurs for the scenario where both options are considered (CCS+NUC),while the highest cost occurs for the scenario with no CCS+NUC.

Figure 4.16 shows that the marginal emission reduction cost in the period 2030-2050 is halvedif CCS is considered. The benefits of having nuclear only are more limited. This can be explainedby the fact that nuclear is an emission reduction option for the electricity sector only, while CCScan be applied more widely. Especially at ambitious emission reduction targets (as is the casehere), emissions must also be reduced outside the electricity sector.

NUC = nuclear

Figure 4.16

Marginal emission reduction cost with various sets of options available

Key point: If CCS is considered, the marginal emission reduction cost is halved

USD

/t C

O2

100

90

80

70

60

40

50

30

20

10

0

2050

2045

2040

2035

2030

2025

2020

2015

2010

2005

2000

CCS+NUC

CCS

NUC

No CCS + NUC

101-124 chapter 4 18/11/2004 07:24

The average emission reduction cost declines by a third if CCS is considered (Figure 4.17). Butthe CCS benefits depend on the availability of nuclear energy. If nuclear is available, the benefitsof having CCS on top of that amount to between 5-10 USD/t CO2 in the period 2030-2050.

NUC = nuclear


Figure 4.17

Average cost of emission reduction with various sets of options available

Key point: If CCS is considered, the average emission reduction cost declines by a third

USD

/t C

O2

50

40

30

20

10

0

2050

2045

2040

2035

2030

2025

2020

2015

2010

2005

2000

CCS+NUC

CCS

NUC

No CCS + NUC

101-124 chapter 4 18/11/2004 07:24

101-124 chapter 4 18/11/2004 07:24

Chapter 5.

CCS SENSITIVITY ANALYSIS

H I G H L I G H T S

■ The use of CCS is not limited by storage constraints or by capture possibilities, but isaffected by the cost of competing technologies (such as renewables) and emission mitigationmeasures (such as land use change and reduction of non-CO2 greenhouse gases), as wellas by policy decisions regarding acceptable levels of climate change risk. Sensitivity analysessuggest that, overall, CCS is a robust option from a cost-effectiveness perspective.

■ For penalties ranging from 10 to 100 USD/t CO2, the CCS potential in 2050 rangesfrom 8 to 25 Gt CO2 per year.

■ At penalties above 15-20 USD/t CO2, the fraction of CCS in total emission reduction isvirtually constant. Under a scenario of optimistic CCS technology assumptions, its sharerepresents over 50% of total emission reduction in 2050.

■ In case of a penalty of 50 USD/t CO2, a 10-15 year delay in putting CCS policies in placeresults in a 20% decline of the cumulative amount of CO2 captured over the wholeperiod 2000-2050. This indicates the importance of timely action. However the quantitiescaptured in later decades are virtually the same.

■ Variations in the price of oil or gas do not significantly affect the CCS potential.

■ Policies that account for a possible decline of conventional oil supply in the long termmay favour more CCS use in combination with synfuel production.

■ The prospects for competing CO2-free electricity options, progress in CO2 capture technology,and the inclusion of developing countries in a global emission reduction effort are keyfactors for the future prospects of CCS.

■ A CO2 penalty limited to OECD countries results in a 53% decline of the CCS potentialin 2050. If commodity trade barriers are completely removed, such a scenario couldresult in industry relocation to countries without CO2 policies. This would result in asubstantial further decline of the CCS potential.

■ The potential of CCS is to some extent influenced by GDP growth, especially by theglobal distribution of this growth.

■ In the case of more optimistic learning assumptions for renewables than in the GLO50reference scenario, the CCS potential declines by up to a quarter. CCS could be consideredas a transition strategy until the full potential of renewables is developed.

■ If nuclear competes on a cost basis alone, the role of CCS could decline significantly,particularly in Japan and the USA. However, such a scenario is unlikely, given that anysignificant expansion of nuclear power in these regions would depend on a change inpublic acceptance and on resolving issues associated with radioactive waste disposal.

5. CCS SENSITIVITY ANALYSIS 125

125-144 chapter 5 18/11/2004 07:27


■ Ultra-supercritical steam cycles (USCSC) and IGCCs with CO2 capture could both play animportant role as efficient and cost-effective generation options for coal. This findingsuggests that CCS is a robust option, independent of the market acceptance of IGCC.

■ Without certain promising but speculative CCS technologies such as IGCC with synfuelcogeneration and chemical looping, total capture declines by a third.

This chapter presents the sensitivity analysis on the GLO50 Scenario that was discussed in Chapter 4. The goal of sensitivity analysis is to quantify the uncertainty that surrounds the ETPresults. Understanding it is a key part of determining if and how CCS should be applied.

The sensitivity analysis indicates that a number of key parameters can significantly impact thepotential of CCS. The variations on these parameters may interact, reinforcing each other or cancellingeach other out. Such interactions are assessed in more detail in the scenario analysis that is presentedin Chapter 6.

It is worth noting that the ETP sensitivity analysis presented here, together with the scenario analysisin Chapter 6, solely concern uncertainties which are within the scope of the ETP model. Certainother uncertainties, such as the legal and public acceptance issues associated with CCS, are outsidethe scope of the model but are discussed in more detail in Chapter 8.

As a starting point, Table 5.1 provides an overview of the parameters used to analyse the uncertaintyin the ETP model results. The table shows the parameter value in the GLO50 scenario and in theuncertainty analysis. Each of these parameters will be considered in turn. The chapter concludeswith an overview of the impact of these parameter variations on CCS use.

Table 5.1

Overview of ETP sensitivity analysis

Variable GLO50 scenario Sensitivity level/range

CO2 penalties 50 USD/t CO2 10, 25, 100 USD/t CO2

CO2 policy scope and timing Worldwide OECD countries only

Policies start 2005 Policies are delayed by 10-15 years

GDP growth and energy demand World average 2000-2050 2.8%/yr World average 2.2%; 3.2% (see annex 3 for regional details)

- 10% additional electricity savings

Nuclear power Growth path fixed in OECD countries Only cost considerations limit growthand growth limited to 5% per year in OECD countries and growthin developing countries limited to 10% per year

in developing countries

Renewables Low learning rates and missing targets High learning rates and ambitiousresult in limited investment policy targets result in morecost reductions cost reductions

Market structure Government guarantees and soft loans, Completely liberalized highlyresulting in low discount rates competitive power industry,

resulting in high discount rates

125-144 chapter 5 18/11/2004 07:27

CO2 Policy Targets

The analysis for the GLO50 scenario in the previous chapter focused in a global CO2 penalty of50 USD/t CO2. This scenario was chosen for detailed discussion because it resulted roughly in anemissions and CO2 concentration stabilization. Future CO2 policies are unclear as they will dependon new insights regarding the urgency of climate policies and on the outcome of a difficult internationalnegotiation process. If and when developing countries curb their emissions is unclear as yet.Consequently, both the penalty level and the penalty scope have been varied. This section discussesthe impact of the penalty level, while the next section discusses the impact of the penalty scope.

Three alternative CO2 policy targets to GLO50 are examined in this section. The penalty levelsstabilize at a level of 10, 25 and 100 USD/t CO2, compared to 50 USD/t CO2 in the GLO50scenario. The impact these levels have on CO2 capture is shown in Figure 5.1. The results suggestthat, even at lower penalty levels, CCS would be a viable alternative on a large scale. This result isimportant because many studies suggest that the damage caused by a tonne of CO2 emissions (interms of environmental impacts, impacts on humans and property) should be valued at less than50 USD/t CO2. Moreover, a significant potential exists for reduction of non-CO2 greenhouse gasesand carbon storage through land use change, at cost levels well below 50 USD/t CO2.

Even at a penalty of 10 USD/t CO2, the amount of CO2 captured reaches 8.4 Gt by 2050. This islikely to represent an overestimation. The model does not account for variations in reservoir geologyand in site-specific CO2 supply and demand within regions that are of particular importance for EOR;it may be that CO2 sources and potential EOR sites are too far apart to allow CO2 use for EOR. Moreover,EOR benefits will depend on the oil yield per tonne of CO2, which is highly site specific. At higherpenalty levels, aquifer storage dominates, which is less sensitive to such site specific factors.

CO2 emissions under various penalty levels are shown in Figure 5.2. The GLO50 scenario is in linewith a stabilization of global CO2 concentrations at a level of around 550 ppm during the 21st

century. Under the GLO10 and GLO25 scenarios, emissions would continue to rise to higher levelsthat are not in line with long-term stabilization at 550 ppm. The GLO100 scenario is the onlyscenario where global emissions decline below 2000 levels. This scenario would be in line with a‘green’ 450 ppm scenario. Note that the 550 ppm and 450 ppm curves start in 2000 at a higher


Variable GLO50 scenario Sensitivity level/range

Technology progress IGCC ‘FutureGen’ and other synfuel IGCC ‘FutureGen’ and othercogeneration, SOFC+CCS and chemical synfuel cogeneration, SOFC+CCSlooping reactors are available and chemical looping reactors

are not considered

CO2-EOR is the only viable EOR option A wide range of competing EORoptions are available

Aquifer storage No aquifer storage

Fuel prices Average oil price: OPEC supply curve twice as steep,2020-2040 30 USD/bbl resulting in higher oil prices

Gas price: 3-5 USD/GJ Gas price: 2-4 USD/GJ

Analysis time horizon 2050 2070

125-144 chapter 5 18/11/2004 07:27


Figure 5.1

CO2 capture at various policy incentive levels

Key point: Higher penalties result in increased CCS use

Gt

CO

2/y

r

25

10

5

0

20

15

2010

2000

2020

2030

2040

2050

GLO50

GLO100

GLO10

GLO25

emission level than the model runs. The difference is accounted for by CO2 emissions fromdeforestation, which is not accounted for in the model. The comparison of model emission projectionsand long-term CO2 concentration stabilization scenarios shows that it is not possible to define aunique target emissions path for the coming decades on scientific grounds. Policy makers shouldaim for emission reductions that balance cost and the risk of important climate change impacts,based on the uncertain information that is available.

Note that in Figure 5.1 the use of CCS keeps rising if the penalty is increased from 50 to 100 USD/t CO2. This suggests that the technical potential is even higher, and the use of CCS is notlimited by storage constraints or by capture possibilities, but by the cost of competing emissionmitigation measures and by policy decisions regarding acceptable levels of climate change risk.

The captured quantities of CO2 shown in Figure 5.1 are high. At the higher penalty levels in 2050they equal current global CO2 emissions. However, these are scenarios with rapid base case emissionsgrowth, ambitious CO2 policies, and limited other options to mitigate emissions. The followinganalyses will show that CCS use would be much lower under slightly different assumptions,but it would still be on a Gt-scale.

Figure 5.3 shows the share of CCS in total CO2 emission reduction. This takes account of additionalemissions caused by energy use for CO2 capture and storage. The share of CCS in total emissionmitigation increases gradually with time, and stabilizes in the period 2040-2050, where it representsover half of total CO2 emission reduction. The penalty level does not have a strong impact on thetotal share of CCS in the emission mitigation.

The results suggest that CCS constitutes a key emission mitigation option at all penalty levelsabove 15-20 USD/t CO2. There is no set threshold value, however, above which CCS should be

125-144 chapter 5 18/11/2004 07:27


Gt

CO

2/y

r

0

10

20

30

40

50

60

70

2000

2025

2050

2075

2100

2125

2150

2175

2200

BASE

GLO10

GLO25

GLO50

GLO100

550 ppm

450 ppm

Source: IPCC, 2001.

Figure 5.2

Energy-related and inorganic CO2 emissions at various policy incentive levels,compared to long-term stabilization scenarios at 550 ppm and 450 ppm

Key point: A penalty of 50 USD/t CO2 is in line with a long-term target to stabilize CO2 concentrations at 550 ppm

considered by policy makers. This is a result on the global aggregation level; such penalty thresholdsmay exist on a regional or sectoral level.

CO2 policy scope and timing

In order to asses the impact of the policy scope (i.e., the regional distribution of CO2 emissionreduction efforts) the GLO50 scenario is compared to a scenario where only OECD countries introducea CO2 policy. Two cases have been analysed. In the first case (OECDHT), the current commoditytrade barriers (tariff and non-tariff barriers) stay in place. In the second case (OECD50), thesetrade barriers are completely removed, and the sum of production cost and transportation cost fromproduction site to the markets determines the industry location choice. The second case is particularlyrelevant as global trade negotiations are aiming for more liberalized markets. This is also theassumption that has been applied in the GLO50 scenario.

Carbon leakage is defined as an increase of emissions in regions without CO2 policies as a resultof CO2 policies in other regions that have introduced CO2 policies (Kuik and Gerlagh, 2003). Onereason for leakage is the relocation of industries to regions without CO2 policies as a consequenceof production cost advantages. Another reason is a redistribution of primary energy use, wherescarce CO2-free (biomass) or low-CO2 (natural gas) energy carriers are increasingly used in regionswith CO2 policies, while the other regions rely increasingly on coal. As a consequence of suchchanges, the potential for CCS declines in the regions with CO2 policies.

125-144 chapter 5 18/11/2004 07:27

CO2 capture in both model runs is compared in Figure 5.4 (the two lower curves). In the GLO50scenario, total CO2 capture worldwide increases to 18.4 Gt in 2050. Almost half of this is capturedin OECD member countries (8.6 Gt CO2). In the OECDHT case, capture in OECD countries is equalto the capture in these countries in the GLO50 scenario. The fact that developing countries anda number of transition economies do not introduce CO2 penalties does result in a decline ofCCS use by 53% in 2050. In the OECD50 model run, the total worldwide CO2 capture potentialdeclines to 3.1 Gt CO2 in 2050. This represents an 83% decline. The difference with the OECDHTcase is that capture in OECD countries also declines by 65% to 3.1 Gt CO2. This decrease can beattributed to so-called ‘leakage effects’.

The future importance of trade barriers is not clear, as it depends in part on any new World TradeOrganisation (WTO) trade agreements coming into force. The results suggest that the interaction oftrade liberalization and CO2 policies is an issue that deserves more attention. One way to prevent majorleakage, with or without trade barriers, is to reach agreement on global CO2 emission mitigation policies.

Another factor which has not been taken into account is that limiting CCS to OECD countries resultsin less learning-by-doing and, therefore, reduces the potential for CCS cost reductions from technologylearning. However, as discussed in Chapter 3, innovation offers the highest learning potential for CCStechnology, meaning that the impact of policy scope on CCS cost reductions is likely to be limited.

In order to assess the policy timing issue, two model runs in particular were analysed. In the firstmodel run, the GLO50 penalty path was followed as discussed previously, but the introduction ofthe penalty was delayed by 15 years. Cumulative capture in the delayed policy case scenario forthe whole period 2000-2050 is about 50-75 Gt CO2 lower than in the GLO50 scenario, equivalentto a decline of about 20%. This shows the importance of timely action. With this delay, capture


Figure 5.3

Share of CCS in total CO2 emissions mitigation

Key point: The share of CCS in total CO2 emissions mitigationdoes not depend on the penalty level

Shar

e o

f em

issi

on

miti

gat

ion

(%

)

60

20

10

0

40

30

50

2010

2000

2020

2030

2040

2050

GLO50

GLO100

GLO25


is 1-2 Gt lower than in the GLO50 scenario, and more in line with the GLO25 scenario up to 2040.After 2040, capture is significantly higher than in the GLO25 scenario, but still below the capturein the GLO50 scenario. The fact that there is a significant capture by 2050, even in the case ofpolicy delays, points to the robustness of the CCS option.

In the second model run, the GLO25 penalty path is followed, but with a delay of 10 years. Delayin the GLO25 case has virtually no impact on the quantities captured. In conclusion, a 10-15 yearpolicy delay has no dramatic impact on whether or not CCS or competing emission mitigationoptions are chosen in the future. The quantities captured in later decades are virtually thesame, although cumulative capture over the whole period 2000-2050 is reduced. Assuming thatone third of the cumulative emission stays in the atmosphere, such a delay would result in atmosphericCO2 concentrations that are about 15 ppm higher in 2050, compared to the GLO50 scenario.

GDP Growth and Energy Demand

Since energy use and CO2 emissions are closely related to economic activity, GDP growth is clearlyan important driver for CO2 emissions. The higher the CO2 emissions, the higher the penalty thatis needed to meet a certain emission target. At a given penalty level, there are also more opportunitiesfor CCS as the emissions from point sources will be higher. Moreover, higher GDP rates imply more

Figure 5.4

CO2 capture for a 50 USD/t case with global policy targets and OECD policytargets

Key point: Restricting CO2 penalties to industrialized countries results in a limited uptake of CCS

Gt

CO

2/y

r

20

4

2

0

10

6

16

18

14

8

12

2010

2000

2020

2030

2040

2050

OECD GLO50 and OECD HT Scenarios

OECD, GLO50

World, GLO50

Note: OECD HT high trade cost (reflecting sustained trade barriers).

125-144 chapter 5 18/11/2004 07:27

investment in new capital stock, which results in more opportunities to introduce CCS. CCS doesnot necessarily increase linearly with GDP, however, as resource availability, for instance, is independentof GDP.

Figure 5.5 shows global GDP and CCS use in the GLO50 scenario with both lower and higher GDPgrowth levels. GDP in 2050 is 25% lower or higher than in the GLO50 scenario. The CCS use is16% lower and 14% higher, respectively. So CCS use increases and decreases with an increase ordecrease of GDP, but the elasticity of CCS use for GDP is less than 1. This can be explained by theassumed regional GDP growth distribution. The global average GDP growth difference is mainlyaccounted for by developing countries (see Annex 3). CCS use per unit of GDP in developing countriesis lower than in OECD countries. Therefore, a higher GDP growth results in lower CCS use per unitof GDP. The policy conclusion that can be drawn from this analysis is that the relevance of CCSfor global emission mitigation is affected by the regional distribution of GDP growth.

A related topic is the impact of future energy demand on CCS use. This is not only a function ofGDP, but it depends also on the energy intensity of GDP and the market uptake of end-use energyefficiency measures. A model run was undertaken in which efficiency in electricity use was increasedover time compared to the GLO50 scenario. This was done by reducing electricity demand by 10%for all regions, and by assuming higher electricity transmission efficiency.

A reduction in electricity demand results in lower electricity production and therefore reduced CCSpotential. The use of CCS may scale with electricity sector investments, rather than electricityproduction capacity. Therefore a 10% decline in electricity demand may have a much more significantimpact on electricity sector investments, and therefore on CCS potentials.


Figure 5.5

The impact of GDP growth on the global use of CCS

Key point: CCS use increases with GDP

Note: Index high GDP growth case 2050=100.

Ind

ex (

-)

2050

2045

2040

2035

2030

2025

2020

2015

2010

2005

2000

0

20

40

60

80

100

GDP, low GDP growth

GDP, ref

GDP, high GDP growth

CO2 capture, low GDP growth

CO2 capture, ref

CO2 capture, high GDP growth

125-144 chapter 5 18/11/2004 07:27


The results show a decline of 25-35% in the use of CCS in the period 2015-2025. In later years,CO2 capture is reduced by 10-15%, compared to GLO50. This pattern can be explained by the factthat in the short term investments are reduced significantly, whereas in the long term all powerplants must be replaced, so CCS potentials are proportional to electricity use. This result suggeststhat demand-side efficiency measures can have a significant impact on CCS potentials in themedium term (2020-2030), but that long-term potentials (2050) are less sensitive.

This result for electricity demand could be extended to other types of final energy use. However,efficiency measures for other types of final energy use are often more costly than for electricity. Arecent study for Germany with a similar model shows that ambitious emission reduction targetswould result in a significant CCS uptake. Without CCS, costly building-insulation measures wouldbe needed to meet the same targets (Markewitz et al., 2004). This result may be country specific,and more analysis is needed on the competition of CCS and energy efficiency measures.

Renewables

The future CCS potential is not only a function of the cost of CCS technologies, but also a functionof the cost of competing emission reduction options. One competing emission reduction option isthe increased use of renewables. Key questions for renewable energy relate to their potential andfuture cost. The cost of renewable energy is largely determined by capital cost, as the primary energyis usually available for free. Future capital costs, in turn, are a function of current capital cost andthe cost reduction that can be achieved through technology learning. This concept was introducedin Chapter 3, where it was discussed for CO2 capture technologies.

The relevance of technology ‘learning by doing’ is much higher for certain renewable energytechnologies than for CO2 capture technologies. Technology learning will be a main mechanism toreduce the cost of renewable energy. This cost reduction, in turn, will affect the competition ofrenewables and fossil fuels with CCS, especially in electricity production. Therefore this sectionfocuses on the sensitivity of CCS for renewables technology learning. This is done through a set ofmodel runs where investment costs for renewables are reduced through increasingly optimistictechnology learning assumptions. In one scenario this learning is based on active governmentpolicies that are under consideration (GLO50REN), in two other scenarios more learning is achievedthrough even higher market uptake of renewables than in the GLO50REN scenario (GLO500515and GLO500718, respectively).

Although cheap renewables options exist today, their potential in terms of energy supply is limited.Future potential and cost must, therefore, be considered in tandem. A detailed GeographicalInformation System (GIS) has been developed for this purpose (see Annex 1).

In the GLO50 scenario, investment cost reductions are based on the cumulative capacities thatfollow from the deployment path in the World Energy Outlook (WEO) Reference Scenario (IEA2004a), in combination with a technology learning rate. Investment costs decline by a fixed fractionfor each doubling of the installed cumulative capacity (IEA, 2000). In the sensitivity analysis(GLO50REN), current and prospective policy deployment initiatives are added to the model asrenewables quota (lower bounds). These deployment targets are defined per renewable energy type.It is possible to quantify what such targets will look like up to the year 2020. It is assumed thatafter 2020, there will be no new policy initiatives and, therefore, no deployment targets are added

125-144 chapter 5 18/11/2004 07:27

(PJ/

yr)

AFR

AU

SC

AN

CHI

CSA

EEU

FSU

IND

JPN

MEA

MEX

OD

ASK

OU

SAW

EU

Prod

ucti

on 2

000

Smal

l hyd

ro3.

01.

760

.716

4.8

52.0

43.4

9.2

3.0

3.5

0.6

1.3

3.7

0.2

34.7

173.

4

Biom

ass

0.0

6.0

26.5

7.146

.62.

39.

24.

846

.60.

01.

65.

51.

425

8.2

176.

5

Geo

ther

mal

1.6

10.5

0.0

0.0

6.8

4.8

0.2

0.0

12.1

0.0

21.2

51.4

0.0

52.6

17.5

Win

d0.

30.

60.

90.

20.

80.

00.

05.

70.

40.

00.

00.

10.

120

.380

.3

Sola

r0.

00.

10.

00.

00.

10.

00.

00.

00.

00.

00.

10.

10.

21.

90.

4

Tida

l0.

00.

00.

10.

00.

00.

00.

00.

00.

00.

00.

00.

00.

00.

02.

1

Targ

ets

2010

(PJ

)

Smal

l hyd

ro4.

82.

774

.022

1.4

84.8

70.6

14.9

4.9

4.7

0.8

1.6

6.6

0.3

42.3

211.

4

Biom

ass

0.1

8.8

39.2

18.3

120.

814

.219

.87.

975

.90.

02.

614

.22.

338

2.2

287.

5

Geo

ther

mal

5.0

13.2

5.7

0.6

14.7

7.8

0.5

0.0

16.2

0.0

45.9

83.7

0.0

70.0

25.9

Win

d9.

52.

023

.71.

028

.44.

74.

714

.71.

20.

00.

10.

20.

212

5.9

324.

9

Sola

r0.

20.

41.

10.

20.

80.

10.

01.

20.

10.

20.

60.

50.

711

.83.

2

Tida

l0.

00.

10.

10.

10.

00.

00.

00.

00.

10.

00.

00.

00.

00.

10.

0

Targ

ets

2020

(PJ

)

Smal

l hyd

ro7.

94.

490

.229

7.6

138.

111

5.1

38.7

7.9

6.3

1.2

2.0

11.8

0.4

51.5

257.

7

Biom

ass

0.4

14.3

58.1

74.2

375.

188

.142

.812

.912

3.6

0.0

4.2

36.7

3.8

622.

546

8.3

Geo

ther

mal

15.6

16.2

5.7

14.2

38.2

12.6

2.2

0.0

21.8

0.0

99.0

112.

50.

014

0.0

46.4

Win

d38

.36.

123

.75.

939

1.3

65.2

18.9

38.2

4.9

0.1

0.5

0.6

0.4

509.

53,

026.

2

Sola

r1.

32.

78.

01.

25.

00.

20.

011

.00.

81.

84.

01.

92.

210

9.8

16.7

Tida

l0.

10.

10.

10.

90.

10.

00.

00.

00.

10.

00.

01.

90.

09.

547

.3


Table

5.2

Ren

ewab

le e

lect

rici

ty d

eplo

ymen

t ta

rget

s in

th

e se

nsi

tivi

ty a

nal

ysis

GLO

50

REN

(PJ

elec

tric

ity/

yr)

125-144 chapter 5 18/11/2004 07:27


to the model after 2020. However, investment costs decline further due to technology learning.Actual investments after 2020 are determined by the model, based on cost-effectiveness.

Deployment targets up to 2020 are shown in Table 5.2. They are based on a wide range ofnational and regional policy plans and policy instruments that have been converted into a singleunit (PJ electricity output per year). Deployment targets have been defined for solar photovoltaics(PV) and thermal solar separately. Those for wind are defined separately for onshore and offshore.These splits are not shown in Table 5.2.

The technology learning rates used in this analysis are listed in Table 5.3. A combination of learningrates and cumulative capacities yield investment cost reductions. The model is based on theassumption of global learning, whereby new capacity anywhere in the world contributes to technologycost reductions in all other world regions. As the installed capacity increases, the investment costsper kW decline. The most important cost reduction occurs for PV, where investment costs in theGLO50REN case decline by 77% between 2000 and 2050. Technology learning has also beenassumed for operation and maintenance costs, but the learning rates are generally lower.

The decrease in future investment cost due to learning-by-doing depends on both investments and learningrates. Learning rates are a source of uncertainty. Factors that commonly complicate their accurate estimationare: new technologies with little or no price/cost history (e.g., PV, fuel cells); technologies with highlysite-specific installation costs (e.g., hydropower, biomass, geothermal); and technologies where marketdynamics obscure the relation of capacity and investment cost (e.g., PV, combined cycle gas turbines).The learning rates used in this study are in line with the range found in the literature (Cody and Tiedje,1997; Neij, 1997; Harmon, 2000; IEA, 2000; Junginger et al., 2005).

The GLO50REN analysis shows about 10% lower CO2 capture than the GLO50 scenario. However,the impact differs by region. In particular, the USA and Europe are affected by lower investmentcosts for renewables. CO2 capture in the USA is about 20% lower, while CO2 capture in Europe is40% lower.

Table 5.4 shows the electricity production mix. Wind is significantly higher in 2030 and 2050, whilesolar is higher in 2050. Electricity production based on fossil fuels with CCS is 21% lower in 2050.The decline for biomass is caused by the reduced opportunities for co-combustion in coal-fired powerplants with CCS.

Table 5.3

Learning rates and investment costs used in the ETP model

Learning rate GLO50 (USD/kW) GLO50REN (USD/kW)(%) 2000 2020 2050 2020 2050

Wind onshore 7 1,000 849 773 684 623

Solar PV 18 5,500 3,196 2,283 1,778 1,270

Solar thermal 5 2,400 2,086 1,912 1,792 1,643

Geothermal 5 1,440 1,378 1,291 1,330 1,245

Small hydro 5 2,500 2,428 2,323 2,392 2,289

Biomass IGCC 10 2,500 2,468 2,373 2,468 2,373

Tidal 5 3,200 2,968 2,780 2,503 2,344

125-144 chapter 5 18/11/2004 07:27


In a second set of sensitivity analyses, the learning rates and the cumulative capacities were morewidely varied for wind and PV.1 A certain minimum quantity of renewables was forced in via alower bound. The investment costs were calculated from a combination of the resulting cumulativecapacities and the learning rates. The model is free to invest more, but this does not result inadditional cost reductions per unit of capacity.2 Four levels of policy targets have been consideredfor PV, and three for wind. These are reflected by minimum capacity constraints. Two technologylearning rates have been considered: 5 and 7% for wind, and 15 and 18% for PV, respectively.

Table 5.4

Electricity production by fuel and technology categoryfor various learning assumptions for renewables, 2050

(EJ/yr) 2030 2050

GLO50 GLO50 GLO50 GLO50 GLO50 GLO50 GLO50GLO50 REN 0515 0718 REN 0515 0718

FF w/o CCS 19.9 15.7 26.5 24.7 20.3 16.4 25.5 21.4

FF with CCS 21.4 16.1 16.9 15.4 56.1 44.2 38.3 32.0

Nuclear 10.5 10.0 9.7 9.6 9.6 9.5 9.4 9.3

Hydro 21.0 20.6 20.9 21.0 24.1 23.5 23.5 21.7

Bio/waste 12.1 14.0 7.9 7.7 15.9 15.8 9.6 9.9

Geothermal 5.7 5.7 7.7 6.9 8.5 7.4 7.5 6.4

Wind 10.9 19.4 12.4 16.5 18.7 31.6 26.7 33.6

Tidal 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0

Solar 0.0 0.7 0.8 2.3 0.0 6.2 12.8 21.5

Total 101.4 102.3 102.8 104.1 153.3 154.5 153.2 155.7

Table 5.5

Cost reductions for PV as a function of the target level

Target Cumulative capacity 2000 15% LR 2050 18% LR 2050

2050 (GW) (USD/kW) (USD/kW) (USD/kW)

0 0.3 5,500 5,500 5,500

1 38.3 5,500 1,817 1,422

2 116.7 5,500 1,400 1,034

3 281.5 5,500 1,138 804

4 654.1 5,500 934 631

LR = learning rate (the investment cost reduction per doubling of the cumulative capacity).

1. PV was selected for this sensitivity analysis instead of solar thermal (concentrating solar technologies) because the cost reductionpotential is more significant.

2. In fact, the model runs are valid ‘least-cost’ solutions only in the case where if the model chooses to invest more than the specified minimum.

125-144 chapter 5 18/11/2004 07:27


The cumulative capacities and resulting investment cost reductions at various learning rates areshown in Tables 5.5 and 5.6. Higher investments and higher learning rates result in lower investmentcosts per unit of capacity. Comparison of investment costs per unit of capacity in 2050 in Tables5.3, 5.5 and 5.6 shows that GLO50REN is situated between levels 1 and 2 for PV, and betweenlevels 2 and 3 for wind.

Any amount of renewables can be forced into the model through model constraints. Within theoptimization framework, the solution is a better systems configuration if the systems costs are lowerthan in a run without such constraints (but with higher investment cost per unit of capacity). Themodel results show indeed that if a CO2 penalty is introduced, most combinations of PV and windtargets result in a reduction of the systems cost. This reduction can be attributed to the moreoptimistic assumption regarding investment cost reductions due to learning, which was not consideredin the GLO50 model run. Not only does this reduction in investment cost per kW reduce the costsof the total investments that were already taking place in the GLO50 scenario, but it allows theintroduction of more renewables at lower investment cost levels as well.

With high learning investments (level 4 for PV and level 3 for wind), the share of renewables increasesto 60% by 2050. The impact on CCS and renewables use in electricity production is outlined inTable 5.4. This very high share of renewables is surprising as it is often stated that intermittencyproblems would limit the share of renewables. However, intermittency is accounted for in the ETPmodel (see Annex 1). Nevertheless, since in reality the model regions nowadays often containmultiple separate electricity grids, the problem of intermittency may be underestimated. Analysison a more detailed level of individual grids is needed to assess this problem in more detail.

With the learning and policy assumptions for renewables in the scenario analysis, fossil fuels withand without CO2 capture and storage would represent 21 to 25% of total electricity productionby 2050, compared to 37% in the GLO50 scenario.3 Total CCS use declines to 15.0 Gt in theGLO500515 case (-18%) and to 13.8 Gt CO2 in the GLO500718 case (-25%).

The results suggest that, compared to GLO50, the future use of CCS would decline by up to aquarter if significant learning effects for renewables occur. However, in all cases CCS can playan important role, and can allow a gradual transition to renewables in the long term. CCS andrenewables should therefore not be considered as competing but as complementary options.

This analysis is based on an extrapolation of past learning effects. Whilst this is a widely appliedapproach, it results in optimistic projections of future cost. A better understanding of the mechanisms

Table 5.6

Cost reductions for wind as a function of the target level

Target Cumulative capacity 2000 5% LR 2050 7% LR 2050

2050 (GW) (USD/kW) (USD/kW) (USD/kW)

0 11.6 1,000 1,000 1,000

1 167.2 1,000 821 756

2 671.6 1,000 741 654

3 3,131.7 1,000 661 557

LR = learning rate (the investment cost reduction per doubling of the cumulative capacity).

3. This excludes biomass co-combustion in fossil-fuelled power plants with CCS, which would add 2-3 percentage points.

125-144 chapter 5 18/11/2004 07:27


that drive future cost reductions is needed in order to reduce the uncertainty. This would enhancethe quality of the policy advice and allow for a better comparison between CCS and renewables.An added advantage of a renewable energy development is that it may still be attractive even ifthe urgency of CO2 emission reductions turns out to be low. This advantage has not been takeninto account in this assessment. It could warrant a preferential treatment of renewables, even ifthe costs are higher. But given the scale of the emission reduction challenge, such a considerationwill not influence the conclusion that both CCS and renewables should be further developed.

Nuclear Power

In the ETP sensitivity analysis, nuclear energy investments are allowed unconditionally in OECDcountries, while the maximum growth rate in developing countries is increased from 5% per yearto 10%. The electricity production mix in the scenario with nuclear is shown in Figure 5.6. Theshare of fossil-fuelled power plants with CCS declines from 39% to 19% in 2050. If cost optimizationwas the only constraint, nuclear would be an attractive option for reducing CO2 emissions.This is based on investment costs that decline from 2,200 USD/kW in 2000 to 2,000 USD/kW in2040 (see annex 1). The results show a significant increase of the nuclear capacity in the worldelectricity mix, up to about a third of total electricity production.

Figure 5.6

Electricity production, GLO50nuclear

Key point: If cost were the only consideration in the selection of emission reduction options, nuclear investments would cut the

share of fossil fuels with CCS in the electricity mix in half

160

120

100

80

140

60

40

20

0EJ e

lect

rici

ty/y

r

GLO50

nuc

lear

GLO50

GLO50

GLO50

nuc

lear

Other

Nuclear

FF with CCS

205020302000

Note: Index high GDP growth case 2050=100.

125-144 chapter 5 18/11/2004 07:27

As illustrated in Figure 5.7, CCS declines by 3.9 Gt (47%) in 2030 and by 7.8 Gt (42%) in 2050if nuclear is considered as an option that competes on a cost basis. The main changes occur inJapan, the USA and Europe where CCS use declines by 75%, 56% and 53%, respectively underthe GLO50nuclear scenario. Clearly, this illustrates that the future role of nuclear energy is of keyimportance for the future role of CCS, especially for industrialized countries.

However, the other three non-cost issues (public acceptance, waste treatment and proliferation)must be solved before a worldwide nuclear renaissance is likely.

CCS Technology Progress

The future role of CCS depends not only on the general energy system characteristics, but also onthe characteristics of CCS technologies. The substantial contribution CCS can make to reducingemissions, as outlined in this chapter, is based on technology that is not yet proven on a commercialscale. If only proven technologies are considered, the role of CCS would be less prominent. Thegoal of this sensitivity analysis is to quantify this decline.

Therefore, in a sensitivity analysis the most speculative technology options have been removed. Giventhe CCS technology mix in the GLO50 scenario, IGCC for cogeneration of electricity and synfuelswas considered a prime candidate for such sensitivity analyses. In a sensitivity analysis, this optionwas therefore excluded. Then in a second, more restrictive sensitivity analysis, all speculative CCStechnologies were removed. In addition to IGCC cogeneration plants, this involved chemical loopingreactors for gas and coal, and power plants including solid oxide fuel cells for gas and for coal.


Figure 5.7

CO2 capture by region, GLO50 and GLO50nuclear

Key point: CCS use in Japan and in the USA declines significantly if nuclear competeson a cost basis, and other nuclear issues are not considered

20

15

10

5

0Gt

CO

2/y

r

GLO50

GLO50

nuc

lear

GLO50

GLO50

nuc

lear

Western Europe

USA

Japan

Other

20502030

125-144 chapter 5 18/11/2004 07:27


Figure 5.8 shows the results for CO2 capture when IGCC with synfuel cogeneration is excluded. TotalCO2 capture is 31% lower in 2050. Moreover, the technology mix changes significantly. CO2 capturefrom IGCC is halved, compared to the GLO50 scenario, while CO2 capture from chemical loopingreactors increases.

In the sensitivity analysis where all speculative technologies are excluded, the decline comparedto the GLO50 scenario amounts to 32% in 2050. So the additional decline compared to the sensitivityanalysis that excluded the cogeneration IGCCs only is small. This suggests that the availability ofthese synfuel cogeneration units is a key uncertainty for total CCS use, but the availability ofchemical looping and fuel cells is less important. However, the technology mix changes. In thismodel run, coal-fired ultra-supercritical steam cycles (USCSC) with CO2 capture emerge as animportant technology.

CCS technology includes not only the capture technology, but storage technologies as well. Inorder to asses the impact of storage assumptions, the parameters for EOR and for aquifer storagehave been varied. With regard to EOR an important uncertainty is if CO2 EOR is really the bestEOR option for a certain field. Rigid analysis would require a field-by-field assessment of the oilrecovery potential of each EOR option. Instead in this analysis the oil recovery potential was keptthe same for all EOR options, and only the availability of EOR alternatives was varied. With regardto the aquifer storage potential, a key uncertainty is whether such storage is really permanent. Theresults so far have been encouraging, but some studies assume, for example, that only confinedaquifers are suited for storage, which would reduce the storage potential significantly (see Chapter 3).In a sensitivity analysis, it is assumed that aquifers are not available for storage.

Figure 5.8

CO2 capture without IGCC transportation fuel cogeneration

Key point: CO2 capture in the electricity sector declines by a thirdif IGCCs with synfuel cogeneration are excluded from the generation mix

20

15

10

5

0Gt

CO

2/y

r

GLO50

GLO50

no IG

CC coge

nGLO

50

GLO50

no IG

CC coge

n

Other

Chemical looping

NGCC

IGCC

20502030

125-144 chapter 5 18/11/2004 07:27

When it comes to CO2 storage, the availability of competing EOR options could reduce the potentialuse of CO2 and may, therefore, affect the overall use of CCS. However, the analysis suggests thatthis is not the case. A penalty of 50 USD/t CO2 results in CO2 being available “for free” for EORuse. Therefore CO2 EOR is cheaper than competing EOR options. CCS use is hardly affected bythe availability of competing EOR options. CCS use declines by 4% in 2030 and by 1% in 2050.

Without aquifer storage, total storage is lower. The decline is limited to 10% in 2030 and 14% in2050. However important regional differences occur. For example, in Korea and Japan there is noCO2 capture and storage if aquifers are not available as storage sites. In other regions, there isheavy reliance on depleted gas fields. The matching of these depleted fields and CO2 sources isnot taken into account in this analysis. This may overestimate the cost-effective storage potentialsin a scenario without aquifer storage. The actual importance of aquifers may therefore be higherthan the model analysis suggests.

Market Structure

In order to assess the role that liberalized electricity markets play in the uptake of CCS, a modelrun with liberalized and highly competitive electricity markets was undertaken. The impact of themarket structure has been simulated by varying the discount rates for electricity-sector investmentdecisions. Discount rates are used to mimic investor behaviour in a MARKAL-type model. Thedifference in discount rates compared to GLO50 amounts to 4.5% (Annex 2). Higher discountrates mean that capital-intensive emission reduction options become less attractive, while optionswith high variable cost become more attractive. For example, nuclear has high capital cost, whilethe capital cost of renewable options vary (high for PV, but comparatively low for biomass). Thecapital intensity of CCS options is high, but not necessarily higher than for other supply-side emissionreduction options. As fuel cost and renewables supply curves differ by region, the impact of thediscount rate on CCS use can differ by region. The impact by region is shown in Table 5.7.

In terms of gigatonnes of CO2 stored, global CCS use is about 10% lower with high discount rates.This decrease is not evenly distributed around the world. While CCS use increases in most OECDcountries, it decreases significantly in developing countries (Table 5.7). This means that, on a regionallevel, market structure can be of importance for the future role of CCS. However, the impact of themarket structure depends on the characteristics of competing emission mitigation options.

Fuel Prices

Many modelling studies suggest that fuel prices are a key factor that will determine future emissionlevels of CO2. The choice of CCS technologies is also influenced by fuel prices (Rubin et al., 2004).While fuel prices tend to fluctuate in the short and medium terms, the analysis in this publicationis more concerned with long-term trends. As a result, long-term oil and gas price trends have beenvaried, without paying attention to more extreme short and medium-term fluctuations.

Oil prices in the BASE and GLO50 scenarios are in line with the World Energy Outlook, reaching29 USD/bbl (in USD of 2000) in 2030 (IEA 2004a). However, much higher or lower oil pricesmay occur. In the ETP sensitivity analysis, the impact of higher oil prices has been assessed. Thesupply curve for oil from the Middle East is twice as steep as in the GLO50 calculations. Thisresults in a higher oil price. The extent to which this curve increases the price depends on the


125-144 chapter 5 18/11/2004 07:27


supply curve for other regions, and on substitution options. The sensitivity analysis results show a20-25% increase in the price of oil up to 2050.

In this high oil price sensitivity analysis the increase of CO2 capture amounts to 0.6 Gt in 2030and 1.8 Gt in 2050. These are relatively minor differences (<10%). Therefore, it can be concludedthat the price of oil is of secondary importance for the use of CCS.

In a second sensitivity analysis for gas, it was assumed that unlimited amounts of gas can beproduced in the Middle East at a wellhead price of 0.5 USD/GJ. This assumption results in areduction of OECD gas prices by 1-1.5 USD/GJ. As a consequence, gas use is higher. OECD gasuse increases by 13% in 2030 and by 18% in 2050, compared to GLO50. Global gas use increaseseven faster. As a consequence of this increase in gas use, global CCS use declines by 1% in 2030and by 7% in 2050. The limited impact can be explained by the high cost of LNG supply from theMiddle East to coal-rich regions such as North America and China. Instead of introducing naturalgas, these regions rely on coal with CCS, even at a reduced gas-supply cost level. Therefore, it canbe concluded that the price of gas is of secondary importance for the use of CCS.

Analysis Time Horizon

The ETP model has a time horizon of 2050. Developments that occur after this date are not takeninto account in the optimal investment path calculated by the model. Such short-sightedness canpose a problem if, for example, oil production peaks shortly after 2050. A transition to a verydifferent energy system configuration might take decades, so sensible energy policies should takesuch depletion into account.

Table 5.7

Change in CCS use in a liberalized market (GLO50 liberalized compared to GLO50)

2030 (%) 2050 (%)

AFR 28 7

AUS 2 5

CAN -25 -3

CHI -9 -18

CSA -1 -28

EEU -21 -20

FSU -45 -62

IND -73 -14

JPN -17 -18

MEA -33 -38

MEX -13 -8

ODA -10 -38

SKO 4 -16

USA 10 6

WEU 29 27

125-144 chapter 5 18/11/2004 07:27

If the post-2050 period is taken into account, the investment path in the period 2000-2050 maylook different. For example, for the electricity sector, post-2050 benefits from technology learningfor renewables may result in an increased use of renewables in the period 2000-2050. In order toanalyse such effects, the time horizon has been extended to 2070, while keeping all parametervalues in the period 2050-2070 constant at their 2050 levels.

The results suggest that total CO2 capture is not affected by such a broader time perspective. Theimpact on total CO2 capture in all periods is less than 5%. On a process level, the differences aremore important. There is about 1.5 Gt more CO2 capture from power plants with synfuel cogenerationby 2050, and less capture from cement kilns and coal-fired chemical looping reactors. This resultcan be explained by the ‘running out of oil’ effect, which leads to a need for significant amountsof synfuels in the second half of the 21st century.

Overview of Sensitivity Analysis Results

The sensitivity analysis suggests that the following sets of parameters are of key importance forthe future role of CCS technology, as they affect CCS use by more than 10% (see Table 5.8):

● The urgency of CO2 capture and sequestration and, therefore, the emission mitigation targets;

● The willingness of non-IEA countries to participate in such a scheme;

● The timing of CO2 policies;

● The feasibility of underground CO2 storage, and the feasibility of speculative CO2 capturetechnologies;

● The feasibility of nuclear energy as a competing emission mitigation option;

● Characteristics of competing renewables emission mitigation technologies;

● Future GDP growth;

● Future energy demand.

The following variables are of lesser importance, as they affect CCS use by less than 10%:

● Liberalized, highly competitive markets versus protected markets (shown as differences in discountrates);

● Oil and gas prices;

● Prospects beyond 2050.

It should be kept in mind that other uncertainties which have not been identified may affect theresults significantly. Also, the input data range that has been applied affects the uncertaintyrange outcome.


125-144 chapter 5 18/11/2004 07:27


Table 5.8

Overview of sensitivity analysis results

Note: Figures in brackets indicate the percentage change compared to GLO50.

Variable Sensitivity level/range Delta CCS 2030 Delta CCS 2050(Gt CO2/yr) (Gt CO2/yr)

CO2 policy scope and timing OECD countries only -3.0 (-35%) to -7.4 (-89%) -9.8 (-53%) to -15.3 (-83%)

Policies delayed by 15 years -2.8 (-34%) -1.9 (-10%)

CO2 penalties 10, 25, 100 USD/t CO2 -5.1 to + 1.9 (-61 to +23 %) -10.0 to +5.6 (-54 to +30%)

Nuclear power Unlimited growth in OECD -3.9 (-47%) -7.8 (-42%)countries and higher growth

in developing countries

Technology progress No IGCC for synfuel -2.1 (-25%) -5.7 (-31%)cogeneration

No IGCC for synfuel -2.5 (-30%) -5.9 (-32%)cogeneration,

Chemical looping, SOFC

Competing EOR options -0.3 (-4%) -0.1 (-1%)are available

No aquifer storage -0.8 (-10%) -2.5 (-14%)

Renewables Important cost reductions/ -0.3 to -1.4 (-4% to -17%) -1.8 to -4.2 (-10% to -25%)policy programmes

GDP growth 2.2%-3.2% -0.8 to + 0.2 (-10 to +2%) -3.0 to +2.5 (-16 to +14%)and energy demand 10%+ additional electricity -1.1 (-13%) -2.4 (-13%)

savings

Market structure Completely liberalized -0.6 (-7%) -1.7 (-9%)

Fuel prices OPEC oil supply curve +0.6 (+7%) +1.7 (+9%)twice as steep

Gas price reduction -0.1 (-1%) -1.3 (-7%)0.5-1.5 USD/GJ

Analysis time horizon 2070 -0.4 (-5%) +0.2 (+1%)

125-144 chapter 5 18/11/2004 07:27

Chapter 6. REGIONAL ACTIVITIES AND CCS SCENARIOANALYSIS

H I G H L I G H T S

CCS RD&D trends and needs

■ Global RD&D efforts to develop CCS technologies are growing rapidly. In an optimisticscenario there will be 10 Mt CO2 capture capacity in power plant demonstration projectsworldwide by 2015. A hundred- to a thousand-fold increase is needed up to 2030 to realizethe global potential for CCS identified in this study and for the technology to have asignificant impact on global emissions. The initiatives announced so far are therefore likelyto be insufficient to realize the potential for CCS identified in the ETP scenario analysis.

■ Worldwide, at least 10 new storage projects should be developed on a Mt per yearstorage scale under varying geological conditions, in order to validate the permanenceof storage and in order to develop regulatory protocols. Given that permanence of storageis a sine qua non for a CCS strategy and the cost of such projects are reasonable, theyshould be established in the short term. Aquifer storage in particular needs to be furtherdeveloped because of its potential importance on a global scale.

■ One IGCC demonstration project with CCS has been announced to date, the USA’sFutureGen project. Canada and the EU have announced plans for further demonstrationprojects for coal-fired power plants with CCS, and Australia has established an emissionreduction project funding programme which may include such plants. It remains to beseen which of these projects are actually realized in the coming years. Even if all fourprojects are implemented, more will be needed to adequately cover the full range of capturetechnologies.

■ Worldwide, five coal-fired 250 MW IGCC pilot plants have been built to date. So far,electricity companies have expressed only limited interest in investing in this technology.Ongoing efforts in Europe to develop high efficiency steam cycles, in combination withCCS, may result in alternatives to IGCC with CCS.

■ To date, not a single demonstration project has been planned for 250 MW+ gas-firedpower plants with CO2 capture. There are plans to do so in Norway, but it is not clear ifand when these will be realized. But the technology challenge for gas seems less than forcoal and biomass, and the CCS potentials are lower on a global scale, so this constitutesa lower RD&D priority than coal.

ETP Scenario Analysis Results

■ CCS can play an important role as a future CO2 emission reduction strategy. Scenarioanalysis suggests a global potential of 3-8 Gt CO2 capture by 2030 and 5-19 Gt CO2

capture by 2050, if ambitious CO2 policies are introduced. This suggests that CCS is arobust strategy.

6. REGIONAL ACTIVITIES AND CCS SCENARIO ANALYSIS 145

145-166 chapter 6 17/11/2004 14:15


■ The scenario analysis suggests that the interaction of key parameters has only limitedimpact on the CCS potentials. Policy incentives and regional policy scope are the mainfactors, followed by technology progress.

ETP Regional Results

■ ETP model analysis suggests that CCS is a robust strategy for North America, Europe andAustralia. The quantities that are captured and stored in all four scenarios are substantial.Even in countries and regions where CCS investments would be limited, the fact that coalremains a viable option has strategic advantages;

■ While the CCS potential for CCS in China and India is significant, its realization will dependon technology transfer from industrialized countries and on global efforts to reduce CO2

emissions. Without substantial emission mitigation efforts in these countries, CCS willnot be introduced. CCS only makes sense in the case of high efficiencies in electricityproduction. Given the comparatively low electricity production efficiency in these countries,priority should be given to increased energy efficiency.

■ In the Middle East, revenues from EOR alone would not be sufficient to merit widespreadCCS use. Therefore, the countries in this region should either be part of a global emissionreduction initiative or CO2 could be supplied from other regions for free or at low cost.

The ETP model sensitivity analysis that is outlined in Chapter 5 showed that a number of keyparameters can significantly affect the potential for CCS. These may interact to reinforce theirrespective impacts or, conversely, to cancel out one another’s impact. Scenario analysis is a way ofassessing such interactions. A scenario is defined as a logical combination of parameters that caninfluence the future role of CCS.

This chapter, the third of four sets of quantitative results from the ETP model, discusses and comparesthe results of four scenarios using ETP analysis. Certain results are valid in all scenarios. This suggeststhat the results are robust. When scenarios show different results, such information can be used todevelop hedging strategies that leave room for a more flexible response. The results for CCS in thefour ETP scenarios are first discussed on a global level. This is followed by a discussion of the regionalscenario results. The regional results are then compared against actual and planned RD&D activities.This analysis provides insights for the CCS policy challenges that are discussed in Chapter 8.

Global CCS Scenario Analysis

Four scenarios are defined. These scenarios are characterized by the acronym EFTEP which standsfor Economy, Fuel demand and price, Technological progress, Environment, and Policy co-operation.

145-166 chapter 6 17/11/2004 14:15

These areas constitute five dimensions that are used to characterize the scenarios. Each dimensionis tracked as a plus (+) or a minus (-), meaning that the set of parameter values is positive or negativefor the future use of CCS.

The dimensions are not completely independent. For example, combining high CO2 urgency andpolicies in industrialized countries only has not been considered, because it seems a less likelycombination. GDP growth and the extent of electricity sector opening have been grouped togetherin the first variable for economic conditions. Higher GDP growth is more likely if markets arederegulated. Combining both is ranked as being positive overall in the EFTEP analysis. Future energydemand has been varied primarily through high or low demand-side efficiency gains for electricity(either in line with GLO50, or in each region 10% lower final demand in 2050, plus additionalgains in transmission efficiency).

Technology progress has been modelled through renewables learning assumptions (in line withGLO50 or with more optimistic assumptions for renewables in line with GLO50REN, see Chapter5), nuclear potentials (maximum potential in line with GLO50 or doubled) and feasibility of CCStechnologies (speculative capture technologies considered or not, aquifer storage considered or not).Balanced sets of technology assumptions have been used, with either optimistic or pessimisticassumptions regarding technology progress for CCS and renewables alike. In the case of limitedtechnology progress for CCS and renewables, nuclear is more widely accepted as a strategy of lastresort.

The penalty level has been varied (25 or 50 USD/t CO2) and the policy scope has been varied(worldwide or only industrialized countries).

The goal of this chapter’s scenario analysis is not to assess the full range of conceivable outcomes,but to show a range of likely outcomes and the way in which key variables might interact. Comparingthe scenarios provides insights into the future role of CCS in the global energy system.

The following four EFTEP scenarios have been defined, the characteristics of which are listed inmore detail in Table 6.1:

● Scenario 1: +++++. High economic growth/liberalized power markets (+), limited efficiencygains (+), rapid technological progress (+), high CO2 mitigation urgency (+), CO2 penalty appliedworldwide (+);

● Scenario 2: +- - - - . High economic growth/liberalized power markets (+), high efficiency gains (-), limited technological progress (-), moderate CO2 mitigation urgency (-), CO2 penalty appliedin IEA countries only (-);

● Scenario 3: -++- - . Moderate economic growth/partially liberalized power markets (-), limitedefficiency gains (+), rapid technological progress (+), moderate CO2 urgency (-), CO2 penaltyapplied in IEA countries only (-);

● Scenario 4: -+-++. Moderate economic growth/partially liberalized power markets (-), limitedefficiency gains (+), limited technological progress (-), high CO2 urgency (+), CO2 penalty appliedworldwide (+).


145-166 chapter 6 17/11/2004 14:15


Figure 6.1 shows the growth in CO2 emissions between 2000 and 2050 in the four EFTEP scenarios.The only scenario which results in a real reduction in emissions is the one in which economicgrowth is limited and there is global co-operation to reduce emissions (-+-++).

Emissions can be stabilized at 30-35 Gt CO2 per year, in line with a 550 ppm stabilization scenario,through either low economic growth or global co-operation. However, in the +++++ scenario, aninitial decline in emissions is followed by a strong increase in the period 2030-2050, which suggestsa further increase beyond 2050.

Table 6.1

Characteristics of the ETP model’s EFTEP scenarios

High:● 3.2%/yr averageworldwide GDPgrowth (see Annex3 for regionaldetails)● Liberalizedelectricity sectorwith high discountrates

Limited efficiencygains

Rapid: ● Includesspeculative CCStechnologies● Renewable energytargets result instrong technologylearning

High CO2

mitigation urgency:50 USD/t CO2

Worldwide (seeGLO50 scenario)

High:● 3.2%/yr averageworldwide GDPgrowth (see Annex3 for regionaldetails)● Liberalizedelectricity sectorwith high discountrates

High efficiencygains from lowgrowth electricitydemand

Limited:● No FutureGen-type ofcogenerationplants● No SOFCs orchemical looping● No CO2 capturefrom blast furnacesand cement kilns● Nuclear acceptedas the last resort

Moderate CO2


OECD countriesonly

Moderate:● 2.2%/yr averageworldwide GDPgrowth (see Annex3 for regionaldetails)● Partiallyliberalizedelectricity sectorwith low discountrates


Rapid:● Includesspeculative CCStechnologies● Renewableenergy targetsresult in strongtechnologylearning

Moderate CO2


OECD countriesonly

Moderate:● 2.2%/yr averageworldwide GDPgrowth (see Annex3 for regionaldetails)Partially liberalizedelectricity sectorwith low discountrates


Limited:● No FutureGen-type ofcogenerationplants● No SOFCs orchemical looping● No CO2 capturefrom blast furnacesand cement kilns● Nuclear acceptedas the last resort● No aquiferstorage of CO2

High CO2


Worldwide (seeGLO50 scenario)

EFTEP Scenario 1: Scenario 2: Scenario 3: Scenario 4:+++++ +-- - - -++- - -+-++

Economic conditions which favour CCS (E)

Fuel demand and prices that lead to efficiency gains (F)

Technological progress on the supply side (T)

Environmental policy through CO2 penalty (E)

Scope of CO2 penalty (P)

145-166 chapter 6 17/11/2004 14:15

The highest emissions occur in the +- - - - scenario, where high economic growth is coupled withlimited CO2 emission reduction efforts in OECD member countries, limited CCS technologydevelopment and limited competing emission mitigation options. Under this scenario, emissions in2050 are more than double the levels seen in 2000.

Figure 6.2 shows the rate of CO2 capture between 2030 and 2050 in the four scenarios. Captureranges from 3-7.6 Gt in 2030 and from 5.5-19.2 Gt CO2 in 2050. Three out of four scenarios areat the lower end of this range. Although the range suggests that the potential is significant, atpresent only the order of magnitude can be given. The potentials in all scenarios are sufficient towarrant further development.

Comparing the scenarios -+-++ and +++++ shows that the main difference occurs post-2030 andis linked to the question of whether CCS is widely applied in developing countries. Comparing thescenarios -++- - and +- - - - indicates that high or low economic growth and high or low technologyprogress are of secondary importance for CO2 capture in industrialized countries.

The results suggest that CO2 capture and storage can play an important role in all scenarios,but its use in 2050 may vary by a factor of four, depending on global co-operation to reduceCO2 emissions and on technology transfer. Also, CCS alone is not sufficient to stem the growthof CO2 emissions. It must be combined with other emission mitigation strategies.


Figure 6.1

CO2 emissions in the four EFTEP scenarios (2000-2050)

Key point: CO2 Emission trends range from a significant increase to a limited decline on 2000 levels

Gt

CO

2/y

r

60

40

20

30

10

0

50

2010

2000

2020

2030

2040

2050

EFTEP - + - + +

EFTEP + + + + +

EFTEP - + + - -

EFTEP + - - - -

145-166 chapter 6 17/11/2004 14:15


Figure 6.2

CO2 capture in the four EFTEP scenarios (2000-2050)

Key point: The four scenarios show between 5-19 Gt of CO2

captured by 2050

Gt

CO

2/y

r

20

10

5

0

15

2010

2000

2020

2030

2040

2050

EFTEP + - - - -

EFTEP - + + - -

EFTEP - + - + +

EFTEP + + + + +

Figure 6.3

CO2 capture by technology type in the four EFTEP scenarios (2030 and 2050)

Key point: Half to three-quarters of all CO2 capture is from IGCCs

20502030

+ + + + ++ - - - -- + + - -- + - + ++ + + + ++ - - - -- + + - -- + - + +

20

15

10

5

0CO

2 c

aptu

re (

Gt

CO

2/y

r)

Chemical looping

IGCC

NGCC

Fuel processing

Steam cycles


145-166 chapter 6 17/11/2004 14:15


Figure 6.3 compares CO2 capture by technology type in 2030 and 2050 in the four scenarios. In2030, the use of IGCC is already dominant and responsible for 45-70% of all CO2 captured. Coal-fired ultra-supercritical steam cycles, gas-fired power plants and capture from fuel processing makeup the difference. In scenario +++++, chemical looping and capture in manufacturing also playan important role. The main difference in 2050 is the extent to which IGCC is applied: CO2 capturefrom IGCCs ranges from 2.3-13.7 Gt CO2 per year.

Figure 6.4 compares CO2 capture by fuel type in 2030 and 2050 under the four EFTEP scenarios.In 2030, capture from coal-fired processes represents between 41-55% of total CO2 captured. In2050, it represents 46-72% of total CO2 captured. The remaining fraction is split between capturefrom gas-fired and biomass-fired processes, and to a lesser extent capture from oil fired processes.The differences in terms of CCS fuel shares for 2030 are limited. In 2050, however, major differencesoccur. High CCS use implies a higher share of coal.

Figure 6.5 compares the primary fuel mix in 2030 and 2050 under the EFTEP scenarios. Totalprimary energy use in 2030 ranges from 587-674 EJ and from 677-956 in 2050. Increasing economicgrowth by 1% per year during the period 2000-2050 results in a significantly higher primary energyuse in 2050. The main variation is in the use of coal, which ranges from 91 to 343 EJ in 2050.Higher coal use allows for more CO2 capture.

The share of renewable energy in primary energy use ranges from 20-34% in 2030 and from 19-34% in 2050. This represents a significant increase from 2000 levels, when renewables, includingtraditional biomass, represented 13% of primary energy use. Biomass represents 65 to 72% of totalrenewable primary energy. As wind and solar electricity are accounted for on an electricity output

Figure 6.4

CO2 capture by fuel type in the four EFTEP scenarios (2030 and 2050)

Key point: Half to three-quarters of all CO2 captured is from coal-fired processes

20502030

+ + + + ++ - - - -- + + - -- + - + ++ + + + ++ - - - -- + + - -- + - + +

20

15

10

5

0CO

2 c

aptu

re (

Gt

CO

2/y

r)

Oil

Coal

Gas

Biomass

Other

145-166 chapter 6 17/11/2004 14:15


basis, their role is more important than their share in primary energy suggests (3.8 to 6.0% by2050). The share of fossil fuels in the primary energy mix ranges from 60 to 77% in 2050. Theirshare amounted to 80% in 2000. So according to the model there is a trend away from fossilfuels, but it is not very strong. The 60% share is for the -+-++ scenario, which has almost 10 GtCO2 capture in 2050. So high rates of CCS use can coincide with a high share of renewables. Infact this is the only scenario where global emissions decline.

Figure 6.6 compares electricity output by fuel and power plant type in 2030 and 2050 under thefour EFTEP scenarios. Power plants with CCS represent between 7-15% of global electricity productionin 2030, which increases to between 8-29% in 2050. This excludes CHP plants with CCS whichrepresent another 9% of electricity production in the +++++ scenario by 2050. This result suggeststhat CCS can play an important role in the electricity sector, but its future share in the electricitymix is uncertain and depends on factors beyond the control of CCS RD&D decision makers. Butthe fact that CCS is applied in all four scenarios suggests that it is a robust strategy.

In summary, the scenario analysis suggests that the urgency of CO2 emission reduction (i.e., thepenalty level) and international co-operation to combat climate change (i.e., the scope of the penalty)are the dominant factors that determine the future role of CCS. Next in order of importance comesCCS technology progress. CCS is to a large extent a coal strategy, except in the -+-++ scenario oflow economic growth and low technology progress, where capture from coal processes representsless than half of total capture. IGCC is a key technology in all scenarios. Only in the +- - - - scenario,its share is less than half of total capture in 2050.

Figure 6.5

Primary fuel mix in the four EFTEP scenarios (2030 and 2050)

Key point: The four ETP scenarios show a wide range for future coal use

20502030

+ + + + ++ - - - -- + + - -- + - + ++ + + + ++ - - - -- + + - -- + - + +

1000

800

600

400

200

0EJ/y

r

Nuclear

Renewables

Oil

Gas

Coal

145-166 chapter 6 17/11/2004 14:15

CCS Potentials and RD&D Activities in a RegionalPerspective

This section and the remainder of this chapter provide an overview of CCS activities and policy plansin major world regions, supplemented by ETP scenario analysis results for these regions. Thecomparison of short- and medium-term policy plans and modelling results is designed to showwhether government activities match the potential for CCS.

A comprehensive overview of individual research, development and demonstration activities in thefield of CCS is available online (IEA GHG, 2004). It includes descriptions of approximately 90 projects,including 11 commercial CO2 capture projects, 35 CO2 capture R&D projects, 26 geologic storagedemonstration projects, 74 geologic storage R&D projects and nine deep ocean storage R&D projects.It is beyond the scope of this study to discuss all these projects in detail. Only some key projectsand concepts will be discussed and compared to the potential identified by the ETP model results.

In 2030, capture in industrialized countries (OECD and transition economies) dominates in allfour scenarios. By 2050 capture in developing countries can reach a similar level to that seen inindustrialized nations. However this is only the case in the scenario +++++. In three out of fourscenarios, capture in industrialized countries dominates total capture in 2050.


Figure 6.6

Electricity production by fuel typein the four EFTEP scenarios (2030 and 2050)

Key point: The share of fossil fuels with CCS in theelectricity mix is limited in three out of the four ETP scenarios

EFTEP2050

EFTEP2030

+ + + + ++ - - - -- + + - -- + - + ++ + + + ++ - - - -- + + - -- + - + +

200

160

120

80

40

180

140

100

60

20

0EJ e

lect

rici

ty/y

r

Wind

Hydro

Nuclear

Centralized FFwithout CCS

CHP

Centralized FFwith CCS

Other Renewables

FF = Fossil Fuelled.

145-166 chapter 6 17/11/2004 14:15


The analysis in Chapter 4 suggested that the bulk of the CCS potential is in the electricity sector.Figure 6.7 shows the electricity produced by power plants with CO2 capture by region under theGLO50 scenario. In 2030, North America represents 37% of the electricity production with CCS.By 2050, Chinese CO2 capture in the electricity sector surpasses North American production. By2050, CO2 capture is concentrated in North America (with 5.5 Gt) and China (4.5 Gt). Captureand storage in Europe amounts to 1.5 Gt. The quantities of CO2 captured in Europe (East +West) are smaller than in certain other regions, but they are still significant. The regional distributionof CCS can largely be explained by differences in fuel prices and fuel availability (better access tonatural gas from pipelines in Europe). The following sections provide a break-down of the regionaloutlook.

North America

Both the US and Canada have significant RD&D programmes investigating the potential applicationof CCS. The US programmes are driven by large indigenous coal reserves, and skepticism as towhether other options can provide sufficient emissions reduction. Moreover, hydrogen from FutureGen-type plants could be used as a CO2-free coal derived transportation fuel. This would reduce CO2

emissions and the need for oil imports. The Canadian programmes aim to find out whether it isworth applying CO2 for EOR. Moreover, the extensive oil sand reserves can only be used to theirfull extent if CO2 policies pose no development constraints. Unlike the US, Canada has ratified theKyoto protocol. Canada therefore has an added incentive to cap emissions.

Figure 6.7

Electricity production by power plants fitted withCCS technology, by region (2030-2050, GLO50 scenario)

Key point: North America and China show the strongestuptake of CCS technology in power generation by 2050

70

60

50

40

30

20

10

0EJ e

lect

rici

ty/y

r

2050

2045

2040

2035

2030

2025

2020

2015

2010

2005

2000

FSU

Other

India

China

OECD Pacific

North America

Europe

145-166 chapter 6 17/11/2004 14:15

RD&D for CCS in the US totaled 40 million USD in 2004. More than 70 projects have receivedfunding (McKee, 2004b). These include:

● 15 projects looking at pre- or post-combustion capture of CO2;

● 17 projects investigating CO2 sequestration, including the potential for terrestrial, geologic and some oceanic storage;

● 14 projects to measure, monitor and verify sequestered CO2;

● 9 projects exploring breakthrough CO2 capture and storage concepts;

● 16 basic research projects within the US National Energy Technology Laboratory.

A detailed overview of these projects is provided by Tomski (2003). The US Department of Energy(DOE) aims to reduce the cost increase for CO2 capture to 10% of electricity production costs (Klara,2003). For this purpose, a wide range of new CO2 capture technologies for power plants are beinginvestigated. Two IGCC demonstration power plants are in operation: Tampa and Wabash. A thirdone, the Sierra Pacific plant, is no longer operational. All three plants have a capacity of around250 MW. None is designed for CCS.

The FutureGen power generation project, one of the US’s major planned initiatives for CCS, willoperate at net 275 MW capacity using IGCC technology to produce both electricity and hydrogenwhile sequestering 1 Mt of CO2 per year. The project will cost 950 million USD with internationalpartners expected to contribute 8% of this sum. The plant is projected to be ready in 2012 withtesting planned by 2015 (DOE, 2004). Since the US has opted for IGCC as the future coal-firedpower plant technology, relatively little attention has been paid to advanced steam cycles.

To date, electricity companies have been reluctant to invest in IGCCs. US companies have considerableexperience in long-range CO2 transportation and the use of CO2 for EOR. However, these EORprojects have not been designed for CO2 storage purposes. A wide range of CO2 storage projectsare underway, but they remain relatively small in scale. One injection project is testing a depletedoil field near Roswell, New Mexico, and a saline aquifer storage project is being developed nearHouston, Texas (the so-called Frio project).

An EOR project in Wyoming known as Teapot Dome, named after a nearby rock formation, is currentlyin its preliminary engineering and testing stages. This would store CO2 from a natural gas processingplant that is transported over a distance of more than 500 kilometres. Storage could begin by 2006and last 7-10 years. The site is projected to store at least 1.6 Mt of CO2 a year when fully operational.The storage potential of another site, the Mt. Simon aquifer, is also being studied in more detail.This reservoir potentially has a very large storage capacity.

Seven regional CO2 sequestration partnerships, which began in August 2003, are aimed at developingframeworks for validation and deployment of CO2 sequestration technologies for US regions. Thetwo-year-long Phase I effort is valued at 18.1 million USD, and will be followed by a Phase II whichwill test CO2 sequestration.

On an international level, the US recently initiated the CO2 Sequestration and R&D LeadershipForum (CSLF). The purpose of the CSLF is to make CCS technologies broadly available internationally,including to developing countries and transition economies, and to identify and address wider issuesrelating to CO2 capture and storage. This could include promoting the appropriate technical, political,and regulatory environments for the development of such technologies. Table 6.2 provides anoverview of approved CSLF projects, some of which are subsequently discussed in more detail.


145-166 chapter 6 17/11/2004 14:15


Table 6.2

CSLF international co-operation projects

Name Partners Topic

ARC Enhanced Canada, The objective of this project is to evaluate, from both economic and Coal-Bed Methane United States environmental criteria, a process of CO2 injection into deep coal beds forRecovery Project United Kingdom simultaneous sequestration of the CO2 and liberation (and subsequent

capture) of coal-bed methane.

CANMET Energy Canada The objective of this project is to demonstrate oxyfuel combustion Technology Centre United States technology with capture of a high-purity CO2 stream suitable for recoveryenhanced oil and to provide information for the scale-up, design and operation of (CETC) R&D Oxyfuel industrial and utility plants based on the oxyfuel concept.Combustion for CO2

Capture

CASTOR European Union The objective of this project is to attempt to validate, from process, France economic, legal, and public acceptance perspectives, post-combustion Norway capture and storage of CO2 with a goal of achieving a major cost

reduction in CO2 capture cost.

CO2 Capture Project, Italy The objective of this project is to continue the development of new Phase II Norway technologies to reduce the cost of CO2 separation, capture, and geologic

United Kingdom storage from combustion sources such as turbines, heaters and boilers.United States

CO2 Separation Japan The objective of this project is to evaluate processes and economics for from Pressurized United States CO2 separation from pressurized gas streams with gas separation Gas Stream membranes.

CO2SINK European Union The objective of this project is to test and evaluate CO2 capture and Germany storage in order to better understand the science and processes involved

in underground storage of CO2 and to provide experience for use in development of future regulatory frameworks for geological storage of CO2.

CO2STORE European Union The objective of this project is to demonstrate, as a follow-on to the Norway current Sleipner project, monitoring to track CO2 migration to undertake

additional studies to gain further knowledge of geochemistry and dissolution processes.

Frio Project Australia The objective of this project is to demonstrate CO2 sequestration in an United States onshore underground saline formation in order to verify conceptual

models and monitoring methods, demonstrate that no adverse health, safety or environmental effects will occur, and develop the experience necessary for larger-scale experiments.

ITC CO2 Capture Canada The objective of this project is to demonstrate CO2 capture using chemicalwith Chemical United States solvents, with a goal of developing improved cost-effective technologies Solvents for separation and capture of CO2 from flue gas.

Weyburn II CO2 Canada The objective of this project is to utilize CO2 for enhanced oil recovery at Storage Project Japan a Canadian oil field, including monitoring of CO2 migration within the oil

United States field, with a goal of determining the overall performance and risks in using CO2 for enhanced oil recovery.

145-166 chapter 6 17/11/2004 14:15

In Canada, an association of seven Canadian utilities and coal producers, together with the US’sElectric Power Research Institute (EPRI), has formed the Canadian Clean Coal Coalition to developcoal-fired power plants with low CO2 emissions. The completed Phase I concluded that aminescrubbing (chemical absorption) is the technology of choice, with gasification and electricity andhydrogen co-production also offering most potential. Phase II will conduct detailed studies ofIGCC plants prior to committing funds for demonstration projects (Canadian Clean Power Coalition,2004).

The Clean Coal Coalition plans to carry out two demonstration projects. The first will look at capturingCO2 from an existing coal-fired power plant and is expected to be operational in 2007. The secondproposal provides for the development, construction and operation of a full-scale demonstrationproject by 2012, which will remove CO2 and other emissions of concern from a greenfield powerfacility. Both demonstration projects will cost 766 million USD, partially funded by the Canadiangovernment (NRCAN, 2004).

On the storage side, Canada’s Weyburn project is looking at using CO2 for EOR with special emphasison CO2 storage, monitoring and validation. The project started in 2001. The 32 million USD PhaseI has been completed (Wilson et al., 2004). Phase II, which started in July 2004, is expected toreceive a similar amount of funding and to last four years. The Weyburn project uses CO2 EOR toincrease oil recovery from 34% to almost 50% of the original oil in place. About 5 kt of CO2 perday is taken from a coal gasification plant in North Dakota (USA), and transported over 330 km.At the conclusion of the project, some 19 Mt of CO2 will have been sequestered in the reservoir.

A number of other R&D projects related to CCS are underway in Canada, backed by financial supportin the form of a 15 million USD incentive programme. These include projects to sequester CO2 inoil sands tailing streams, to use CO2 in ECBM pilot projects, to use CO2 to enhance gas hydrateproduction, and an oxyfuel demonstration project. The intention of the US-supported programmeis to stimulate the growth of a Canadian CO2 capture and storage industry. Eligible expendituresare defined as up to 50% of the cost of capital equipment and all other direct expenses requiredfor capturing, compressing, transporting and injecting CO2 (NRCAN, 2004).

Canada could become an important oil supplier if the oil sand production is further developed.However, the production of crude from these sands would be an important source of CO2. Due tothe geology of Canada, the province of Alberta is where current oil, oil sands and gas activities areconcentrated. The provincial authorities are actively supporting research in the development ofCCS technology in order to allow future expansion of these production activities. A total of 11 millionUSD of royalty credits has been approved for four CO2 EOR projects in Alberta.

The fact that important CCS RD&D activities are taking place in North America is in line with theprojected potential importance of this strategy for this region. Figure 6.9 compares CO2 capturefrom major point sources in 2030 and 2050 in North America under the four EFTEP scenarios.Total capture is significant, up to 5 Gt CO2 per year in 2050. Since the amount captured is notvery scenario dependent, CCS therefore seems to be a ‘safe bet’. The results suggest 2-3 Gt CO2

capture by 2030. Electricity production represents the bulk of the CO2 capture potential, while thestorage potential of the FutureGen project is 1 Mt per year by 2015. A 2,000-fold increase is neededin order to realize the storage potential shown in Figure 6.8. Expansion on this scale presents amajor challenge. Although current policy efforts are important, they are not geared to deployingCCS on such a large scale within the next 25 years.


145-166 chapter 6 17/11/2004 14:15


Europe

For the purposes of the discussion in this section, Europe includes Eastern and Western Europe. Atpresent, Europe already has advanced CO2 emission reduction policies in place. These include CO2

market mechanisms, demand-side policies, and support programmes for renewables and otheremission reduction technologies. CCS is gaining increasing attention, as policy makers start to realizethat significant emission reductions require a wider portfolio of emission mitigation strategies.CO2 may also be used for EOR in the maturing North Sea oil fields.

The prospects for CCS differ by country. Norway, for example, is very active in this area in the fieldof subsea aquifer storage through the Sleipner demonstration project and the planned SnohvitLNG project. There is also considerable interest in using CO2 for EOR. CENS and Statoil’s New Energygroup are studying the supply of CO2 for EOR to the Gullfaks field within the Statoil Tampen 2020project (see Chapter 3) (Coleman, 2004). However, the project is not cost-effective without a CO2

incentive. Norway has conducted a number of feasibility studies for gas-fired power plants withCO2 capture, and Denmark has studied the feasibility of CO2 capture for coal-fired power plants,but these studies have not yet resulted in any further demonstration plans.

In Europe as a whole, electricity companies have so far been reluctant to invest in IGCC technology.To date, two IGCC demonstration projects have been built in the region, at Buggenum in theNetherlands and at Puertellano in Spain, each with a capacity of around 250 MW. More interestis hoped for from the EU’s CASTOR project, which began in 2004 and is under the leadership of

Figure 6.8

CO2 capture in North Americain the four EFTEP scenarios (2030 and 2050)

Key point: CO2 storage potential in North America is up to 5 Gt CO2

and supported by high CCS use in all scenarios

20502030

+ + + + ++ - - - -- + + - -- + - + ++ + + + ++ - - - -- + + - -- + - + +

6

3

1

5

4

2

0Gt

CO

2/y

r

Fuel processing

Electricityproduction


145-166 chapter 6 17/11/2004 14:15

the Institute Française de Petrole (IFP). The project involves 30 companies and research institutionsfrom eleven European Union countries and aims to reduce the cost of capturing and separatingCO2 from flue gases to 20-30 EUR/t. Much of the research on capture, representing 70% of thefour-year Castor budget of 19.1 million USD, will focus on a pilot plant able to treat 1-2 t/hr ofCO2 from real flue gases (10 kt/yr or 1% of the US’s FutureGen demonstration plant).

The EU is co-funding various storage projects. One is the first CO2 storage in an onshore aquifer inKetzin, close to Berlin, known as CO2Sink. Previously, the site was used for natural gas storage. Thegoal is to improve understanding of the behaviour of CO2 underground (GFZ Potsdam, 2004). TheRECOPOL project in southern Poland is an EU-funded pilot/demo project for CO2 ECBM. Withinthe CASTOR project, storage research will take place at four sites: in the abandoned Casablancaoil reservoir off Spain, the Snohvit aquifer storage project in the Norwegian Sea, a depleted offshoredeep gas reservoir owned by Gaz de France in The Netherlands (also known as the CRUST project,mentioned below), and a depleted shallow gas reservoir owned by Rohoel-Aufsuchungs AG in Austria.

A large number of CCS projects are being co-funded by the EU. It is beyond the scope of this analysisto discuss all of these in detail. Further information can be found on the IEA GHG project database(IEA GHG, 2004).

The European Commission has announced the so-called Quick-start hydrogen programme. This isa 10-year programme with a budget of 3.4 billion USD (2.8 billion EUR), that covers hydrogensupply and hydrogen demand. From December 2004, demonstration projects can be proposed forhydrogen production from fossil fuels with CCS (HypoGen). A budget of 1.6 billion USD (1.3 billionEUR) is foreseen for HypoGen, roughly half the funding for which would come from the EU frameworkprogrammes. HypoGen focuses on hydrogen and electricity cogeneration and is similar to the US’sFutureGen project. HypoGen aims to demonstrate the economic viability of hydrogen and electricityproduction from de-carbonated fossil fuels, to prove concepts and test the regulatory environmentfor safe and reliable geological storage of CO2 (European Commission, 2004).

The EU Emissions Trading Scheme (ETS) which begins in 2005 will provide incentives for CO2

emission reduction. CCS is mentioned in the relevant ETS Directive, but emission reductions mustbe proven. The permit price under the ETS is projected to be less than 10 USD/t CO2, which willbe insufficient as an incentive for CCS use, but which may help to demonstrate feasibility for certainEOR projects. However, penalties of up to 100 EUR/t CO2 are envisaged for the period 2007-2012for non-compliance. These penalties are much higher than CCS costs, and therefore CCS may beintroduced, if other strategies do not result in sufficient emissions reduction.

In terms of individual countries, Germany has started a large RD&D programme known as Cooretecwhich aims to develop and demonstrate energy efficient fossil-fueled power plants, including CO2

capture technologies (Cooretec, 2003). It includes a roadmap to further increase efficiencies by20% by 2020. An oxycoal process for CO2 separation will be developed. Cooretec funding amountedto 10 million EUR in 2004, and will amount to 30 million EUR per year in 2005 and 2006. Thegoal of the programme is to maintain Germany’s leading role as a power plant supplier. Most workis focused on materials and systems design for high-efficiency steam cycles with attention to IGCCrestricted to desk studies. While these efforts are not directly focused on CCS development, energyefficiency improvements constitute a crucial step that will enable CCS introduction.

Representatives of the regional government of North Rhine-Westphalia have indicated that thedecision to construct a reference coal power plant might be taken before the end of 2004 (Geipel,2004). This would be a hard coal-fired power steam cycle plant with 45.9% efficiency, which canbe increased to 47.3% with additional investments of about 50 USD/kW. A large-scale test facility


145-166 chapter 6 17/11/2004 14:15


for USCSC will be built in the German Scholwen power plant as part of the so-called AD700project (AD700, not dated).

In the UK, the Department of Trade and Industry’s Carbon Abatement Technologies (CAT) programme(DTI, 2004) is developing a strategy for a near-to-zero emission fossil-fuel combustion plant. Thenew CAT strategy is expected to be finalized by autumn 2004. It will address a number of strategicissues that need to be considered before a decision can be taken on the demonstration of CCS. Inthe UK, the fourth call for carbon abatement technologies runs for three years and covers ten projectsfor a total amount of 18 million USD over a period of three years. Topics include combustiongasification, efficiency, emissions, design studies, CO2 capture and hydrogen (Morris, 2004).

The longer-term strategic importance of CCS is recognized in the UK government’s Energy WhitePaper. CO2 EOR receives special attention because the UK fields in the North Sea are rapidly depleting(DTI, 2003). This opportunity only exists in the short term, however, and CO2 injection needs tostart by 2006/8 if it is to have an impact on the largest fields before the existing infrastructure isdismantled.

The Netherlands plans a small pilot project for offshore storage in a depleted gas field, called CRUST(CO2 Re-use through Underground Storage). The Netherlands has also launched the CATO programme(CO2 capture, transport and storage) for further technology development (Tweede Kamer, 2003).The programme funding amounts to more than 30 million USD over a period of five years (2004-2008), half of which is government funding.

Italy has a national CCS R&D programme in which hydrogen production from fossil fuels is a keypriority. Underground storage potentials have already been analysed with two pilot projects planned.

Figure 6.9

CO2 capture in Eastern and Western Europein the four EFTEP scenarios (2030 and 2050)

Key point: CO2 storage potential in Eastern and Western Europeis up to 1.6 Gt CO2, although the potential is halved under some scenarios

20502030

+ + + + ++ - - - -- + + - -- + - + ++ + + + ++ - - - -- + + - -- + - + +

1.8

1.6

0.8

0.2

1.2

1.4

0.6

1.0

0.4

0.0Gt

CO

2/y

r

Fuel processing



145-166 chapter 6 17/11/2004 14:15

One is the Sibilla EOR project in the Adriatic Sea which will sequester 1.5 Mt of CO2 over a periodof 10 years, starting in 2007. The second is an ECBM project in the Sulcis area of Sardinia, where1 Mt CO2 sequestration is planned (Capra, 2004).

Figure 6.9 compares CO2 capture from major point sources in 2030 and 2050 in Western andEastern Europe under the four EFTEP scenarios. Total capture is significant, up to 1.6 Gt CO2 peryear in 2050. The capture potential is half to a third of the potential in North America. Theamount captured is not very scenario dependent, and so, as in the case of North America, CCSseems a ‘safe bet’. The extensive CCS RD&D activities taking place in this region are in line withthe projected potential importance of this strategy for the region. The results suggest 0.4-0.8 Gtof CO2 capture by 2030. Current policy efforts are not geared to CCS use on such a large scalewithin the next 25 years.

Asia-Pacific OECD Countries

The OECD Asia-Pacific region encompasses Australia, Japan, New Zealand and Korea. Since NewZealand has very low CO2 emissions and Korea has no CCS plans to speak of, the following discussionis limited to Australia and Japan.

Australia relies on coal for the bulk of its electricity production. As the largest coal exporter in theworld, Australia has important business interests to develop CCS technology. A technology roadmaphas been published by the Co-operative Research Centre for greenhouse gas technologies in Australia(CO2CRC, 2004). Since 1999, the Australian Petroleum Co-operative Research Centre (APCRC) hascarried out research into deep geological storage of CO2 through its GEODISC programme, whichshows that Australia has very high potential for cost-effective geological storage of CO2.

The Australian and State governments are working with industry to support CCS R&D. The Australiangovernment is spending approximately 20 million USD per year on all clean coal technology research,of which a significant proportion is directed to CCS R&D. This includes design of energy efficientcoal-fired power plants (hard coal and lignite), new capture technologies, CO2 storage, and hydrogentechnologies. The government has established a 350 million USD fund that should attract another700 million USD in private investment to develop and demonstrate low emission technologies,including CCS (Jones, 2004).

Aquifer storage is planned for the Gorgon gas field situated 130 km off the north-west coast ofWestern Australia. CO2 separated for LNG production will be re-injected into the gas field, withapproximately 5 Mt CO2 storage aimed at per year. Production from the Gorgon gas field is projectedto begin in 2008-2010 (Gorgon, 2004).

Japan has studied CO2 capture and storage for quite some time. However, onshore undergroundstorage potentials are limited by the geology of the country and the lack of indigenous oil and gasreserves. As a consequence, studies have focused on oceanic storage, but this strategy is highlycontroversial in Japan and abroad. Therefore, attention is now switching to ECBM, with a 10 Gtcumulative storage capacity, and a pilot project is underway. A small aquifer storage pilot projecthas also started, where 10 kt of CO2 will be injected over a period of one and a half years (RITE,2003).

The Engineering Advancement Association of Japan (ENAA) undertook estimates for geologicalstorage potential in the early 1990s. These estimates indicated the potential to store some 92 GtCO2 in geological reservoirs, the majority of which (52 Gt) are offshore aquifers. Seen in comparison


145-166 chapter 6 17/11/2004 14:15


to 500 Mt CO2 emissions per year from stationary sources, geological storage in Japan wouldseem to have significant potential. However, Japanese storage potentials are not evenly distributed,which will limit the practical storage potential.

In 2001, Japan launched a new research project, involving the Research Institute of InnovativeTechnology for the Earth (RITE) and ENAA, which will build upon the earlier research work. The 5-year project will involve a number of activities, including:

● A field-scale injection study to demonstrate the potential for CO2 injection in Japan and obtaindata on the actual behaviour of carbon dioxide underground.

● A geological survey around the Pacific offshore region of Japan. The study will compile existingseismic and exploration data in the region and generate a GIS database that will act as a supporttool for future storage activities.

● Undertake a system analysis to assess possible combinations among locations of large-scale CO2

sources and storage options. A cost evaluation model will be used to assess cost-effective storageoptions for Japan (Gale, 2002).

So far, Japanese policies are targeted at increasing energy efficiency and the use of nuclear powerto reduce emissions. However, the future expansion of nuclear energy is controversial. As a result,R&D is looking at maximizing coal-fired plant efficiency with most attention focused on fluid bedcombustion. A 25 MW IGCC demonstration project, known as EAGLE (Coal Energy Applicationsfor Gas, Liquid & Electricity) is now underway. This plant integrates fuel cells and could be usedfor synfuel cogeneration (similar to FutureGen). So far, there are no plans for CO2 capture from thisinstallation (Wakamatsu, 2004).

Figure 6.10

CO2 capture in the OECD Asia-Pacific regionin the four EFTEP scenarios (2030 and 2050)

Key point: CO2 storage potential in the OECD Asia-Pacific regionis up to 2.1 Gt CO2, although the potential is halved under some scenarios

20502030

+ + + + ++ - - - -- + + - -- + - + ++ + + + ++ - - - -- + + - -- + - + +

2.5

2.0

0.5

1.5

1.0

0.0Gt

CO

2/y

r

Fuel processing



145-166 chapter 6 17/11/2004 14:15

Figure 6.10 compares CO2 capture from major point sources in 2030 and 2050 in Japan, Korea,Australia and New Zealand under the EFTEP scenarios. Total capture is significant, up to 2.1 GtCO2 per year by 2050. Capture is of a similar order to that in Europe, even though the OECD Asia-Pacific region has only a third of the population. The potential for CO2 capture is significant,especially in Australia. The amount that is captured is not very scenario dependent, so once againCCS seems a ‘safe bet’, similar to the situation in North America and Europe.

The fact that important CCS RD&D activities are taking place in Australia is in line with the projectedpotential importance of this strategy for this region. The results suggest 0.5-0.1 Gt CO2 capture by2030. Current policy efforts are not geared to CCS use on such a large scale within the next 25 years.Electricity production represents the bulk of the CO2 capture potential. Capture in manufacturingindustry and fuels processing is also of importance.

China

China relies to a large extent on coal for its energy supply. A recent study predicts that the efficiencyof coal-fired power plants will increase from 32.0% in 2000 to between 39.2% and 44.4% by 2020(ERI, 2003). This means that new Chinese coal-fired power plants would achieve current OECD best-practice efficiency levels by 2020. However, opinions in China diverge on whether the countryshould adopt advanced steam cycles or IGCC technology for new plant. Approval has been givenfor a feasibility study for a 300-400 MW IGCC plant in Shangdong province. As discussed in Chapter3, either choice does not impede the use of CCS, as long as the plants achieve high efficiencies.

China has considerable potential for the capture and utilization of coal bed methane (CBM). Atpresent, the China United Coal-bed Methane Corporation (CUCBM) remains the sole professionalstate-owned company responsible for CBM exploration, development, production, pipeline constructionand sale in China. In addition, CUCBM has obtained exclusive rights for the exploration, developmentand production of CBM, in co-operation with foreign companies.

Twenty CBM projects with international co-operation have been signed, covering an area of32,000 km2 and representing a reserve of 3,654 billion m3 (more than 100 EJ gas; BHPBilliton,2003). CUCBM is planning two CO2 ECBM field tests in co-operation with a consortium of Canadiangroups (Law and Gunter, 2003). It remains to be seen whether the coal permeability makes ECBMa viable option. As in other oil-producing countries, China is also interested in enhancing the outputfrom oil reservoirs as their output diminishes over time. An enhanced oil recovery project is currentlyunderway at the Liaohe Oil Field, looking at injecting boiler flue gases into a production well (Zhuet al., 2001).

Figure 6.11 compares CO2 capture from major point sources in 2030 and 2050 in China under thefour EFTEP scenarios. The total capture potential is significant, up to 4 Gt CO2 per year in 2050.At this level, capture is of a similar order to that in North America. However, the amount that iscaptured is very scenario dependent, and much lower than in other regional scenarios. The policychoice to mitigate CO2 emissions is a key variable. But even in the -+-++ scenario with a CO2 penaltybut with less optimistic technology assumptions, the use of CCS in 2050 is small, although it isstill at the same level as in Europe. The fact that no important CCS RD&D activities are takingplace in China is in line with the uncertain future of CO2 emission reduction in this region. Chinais participating in the CSLF. The results suggest that more Chinese involvement in CCS mightbe warranted.


145-166 chapter 6 17/11/2004 14:15


India

The Indian energy system will continue to rely on indigenous coal, which is largely high ash coal.In combination with subcritical steam cycle power plant technology, this results in low efficienciesin electricity production. This issue has been discussed in Chapter 3. As a consequence of this lowefficiency, CO2 capture would make little sense for India on the short and medium term. As a firststep, a programme is needed to establish high-efficiency, large-scale power plants. At a later stagethis could be followed by CO2 capture and storage. To date, India has no programmes in the fieldof CCS. However the country is participating in the CSLF. There is some potential for ECBM, butthe cost for CO2 capture would outweigh the benefits of ECBM.

Figure 6.12 compares CO2 capture from major point sources in 2030 and 2050 in India under thefour EFTEP scenarios. The total capture potential is significant, up to 2.5 Gt CO2 per year in 2050.At this level, capture is of a similar order as in Europe. However, the amount that is captured isvery scenario-dependent, and it is negligible in two out of the four scenarios. The policy choice tomitigate CO2 emissions is a key variable. But even in the -+-++ scenario with a CO2 penalty butwith less optimistic technology assumptions, the use of CCS in 2050 is small. The fact that noimportant CCS RD&D activities are taking place in India is in line with the uncertain future of thisstrategy for this region. It would be better for India to focus first on switching to high-efficiencypower plants, with the possibility for retrofitting CCS at a later stage.

Figure 6.11

CO2 capture in China in the four EFTEP scenarios (2030 and 2050)

Key point: CO2 storage potential in China is around 3.8 Gt CO2, although much lowerquantities or none are observed in three out of the four scenarios

20502030

+ + + + ++ - - - -- + + - -- + - + ++ + + + ++ - - - -- + + - -- + - + +

4.5

4.0

2.0

0.5

3.0

3.5

1.5

2.5

1.0

0.0Gt

CO

2/y

r

Fuel processing



145-166 chapter 6 17/11/2004 14:15

Middle East

The Middle East has the energy resources to become the main source for oil and gas in the comingdecades. At the same time, many of its giant oil fields are ageing. For example, half the provedreserves have been produced from Ghawar, the world’s largest oil field in Saudi Arabia (Abdul Baqiand Saleri, 2004). CO2 EOR could be a way of increasing Middle Eastern oil production from fieldsthat are in decline. There are two challenges for widespread CCS use in this region. The first ofthese is of a technical nature, while the second concerns CO2 supply at acceptable cost.

About 80% of oil that is currently produced in the region is medium and heavy sour crude (ENI,2004). Medium crude is crude between 26-35°API (0.898 to 0.845 t/m3). This oil is heavier thanoil in fields where CO2 EOR has been successfully applied so far. Various authors give maximumoil gravities for CO2 miscible floods ranging from 0.8 to 0.95 t/m3, but typically 0.9 t/m3 is considereda maximum (see Chapter 3, Shaw and Bachu, 2002). The oil that remains underground followingwater injection is probably heavier than the oil that is produced at the moment. Therefore thetechnical potential of CO2 EOR needs more attention.

Also, reservoir temperature should not exceed 120°C to ensure CO2 and oil miscibility. This wouldexclude the Ghawar field, for example, which has a reservoir temperature of 137 to 150°C. If miscibleflooding is not possible, immiscible CO2 flooding may be applied. However this will cut the EORoil recovery factor in half. A careful, field-by-field assessment is needed if CO2 injection is to bejudged suitable for recovering significant amounts of remaining Middle Eastern oil.


Figure 6.12

CO2 capture in India in the four EFTEP scenarios (2030 and 2050)

Key point: CO2 storage potential in India is up to 2.6 Gt CO2, but much lowerquantities or none are observed in three out of the four scenarios

20502030

+ + + + ++ - - - -- + + - -- + - + ++ + + + ++ - - - -- + + - -- + - + +

3.0

2.5

1.5

0.5

2.0

1.0

0.0Gt

CO

2/y

r

Fuel processing



145-166 chapter 6 17/11/2004 14:15


While the Middle East has a significant storage potential in depleted oil and gas fields, it is not aprimary emissions source in a region where emissions amounted to 1 Gt CO2 in 2000 (IEA, 2002b).The region’s electricity production is largely based on indigenous oil and gas reserves, not on coal.CO2 could be imported from other regions, however. This could even be a feasible option withoutCO2 policies in the Middle East. For distances up to 5,000 kms (Western Europe-Middle East)transportation would be in the 15-25 USD/t CO2 range (IEA GHG, 2002a; IEA GHG 2004b).These transportation costs must be balanced against the EOR benefits.

Figure 6.13 compares CO2 capture from major point sources in 2030 and 2050 in the Middle Eastunder the four EFTEP scenarios. Total capture is significant at up to 0.35 Gt CO2 per year in2050. However, this potential is an order of magnitude smaller than in the regions discussedearlier, and the amount that is captured is very scenario dependent. Apart from CO2 shippingfrom other regions, a closer look may reveal certain low-cost CO2 supply options, possibly withrelocation of oil or gas-intensive industries with a low-cost CO2-capture potential to the MiddleEast (ammonia, DRI etc.). The analysis in Chapter 4 suggests that such relocation would dependon future commodity transportation cost and trade barriers. The viability of such a strategy needsto be studied in more detail.

Figure 6.13

CO2 capture in the Middle Eastin the four EFTEP scenarios (2030 and 2050)

Key point: CO2 storage potential in the Middle East is around 0.35 Gt CO2,although no potential is observed in two of the four scenarios

20502030

+ + + + ++ - - - -- + + - -- + - + ++ + + + ++ - - - -- + + - -- + - + +

0.40

0.35

0.15

0.05

0.25

0.30

0.20

0.10

0.00Gt

CO

2/y

r

Fuel processing



145-166 chapter 6 17/11/2004 14:15

Chapter 7.

THE IMPACT OF CCS ON ENERGY MARKETS:MODEL RESULTS

H I G H L I G H T S

The analysis of the impact of CCS on energy markets has concentrated on the global level.Country specific results may differ, depending on the resource endowment and trade effects.

Impact of CCS on Coal Markets

■ In the case of CO2 penalties above 25 USD/t CO2 with the CCS option available, coaluse is stable up to 2030, but doubles from 2030 to 2050. The increase in the use of coalis accounted for by electricity and synfuel production. While coal use increases in absoluteterms, it declines in relative terms compared to the BASE scenario projection for 2050.CCS plays a key role in keeping coal a viable option. Without CCS, coal use in 2050 declinesby 64% in a scenario in which a USD 50/t CO2 penalty is imposed, compared to the samepenalty level with CCS.

■ If CCS is not considered, coal prices decline by 10%, compared to the scenario with CCS.This decline can be attributed to lower coal demand. These fuel price impacts are smallcompared to the penalties for emissions caused by coal use without CCS.

Impact of CCS on Gas Markets

■ Gas use doubles between 2000 and 2050. This growth is not significantly affected byCO2 policies and/or the availability of CCS. For example, exclusion of the CCS optionresults in gas use variations of +/- 15%, depending on the period and the penalty level.

■ If CCS is not considered, the impact on gas prices is mixed and region specific, but canbe significant. Modelling results suggest a small price decline for Europe and a significantincrease for the US. As in the case of coal, these fuel price impacts are small comparedto the penalties for emissions caused by gas use without CCS.

Impact of CCS on Oil Markets

■ Oil production declines by 10-20% at higher CO2 penalty levels. If CCS is not considered,an additional 10% decline of oil production occurs at penalty levels above 50 USD/t CO2.At lower penalty levels the impacts of having CCS are small. This assumes that technologyalternatives exist for CO2 enhanced oil recovery.

■ If CCS is not considered, oil prices would increase by about 10%, compared to a scenariowith CCS. As is the case for coal and gas, fuel price impacts are small compared to thepenalties for emissions caused by fuel use.

7. THE IMPACT OF CCS ON ENERGY MARKETS: MODEL RESULTS 167

167-180 chapter 7 18/11/2004 07:31


Impact of CCS on Renewable Energy

■ Biomass is the most important renewable energy option for CO2 policies. Its use cangrow up to 150 EJ in 2050. Biomass use is barely affected by CCS.

■ In the BASE scenario, the use of other forms of renewable energy increases threefoldfrom current levels during the period 2000-2050. If CO2 policies are put in place, a five-to-sixfold increase in renewable energy use (excluding biomass) could occur. Additonally,other renewables use can increase by up to 40%, in cases where CCS is not considered.This additional renewable energy is mainly used in the electricity sector.

■ With more optimistic learning assumptions for renewables, the share of CCS in the electricitymix may decline from 37% to 20%. The share of renewables would increase accordingly.Both renewables and fossil-fuelled power plants with CCS will be needed for an electricitysupply system with low emissions.

Impact of CCS on Electricity

■ Excluding CCS results in a 4 to 52% increase of electricity prices (feed-in prices). As aconsequence of price increases, electricity demand declines by 7% in 2050, compared tothe same scenario with CCS. Without CCS, renewables substitute part of the fossil fuelsin the electricity mix.

This chapter, the fourth and final set of quantitative results from the ETP model, discusses theconsequences of deploying CCS on primary energy (coal, gas, oil and renewables) and electricitymarkets, using ETP model analysis. Apart from the obvious environmental concerns, supply securityand economic factors play an important role in the design of energy policies. While the analysisin the previous chapters has shown that a CCS strategy can significantly reduce CO2 emissionsand also lower the cost of environmental policies, it is less clear what impact CCS would have onfuel markets. This chapter discusses the impacts on fuel quantities and fuel prices.

The effects of CCS on fuel markets can be split into three categories: fuel substitution effects (e.g.,enhanced coal competitiveness compared to other fuels), effects on fossil fuel recovery and increasedenergy demand for CCS.

The additional oil and gas recovery potential due to CO2 use is elaborated in Chapter 3. If CO2 isused for EOR, it increases oil recovery by 10-40% compared to a situation without EOR. Injecting CO2

into depleted gas fields can increase gas recovery by up to 5%. Enhanced coal-bed methane recoveryalso has significant gas supply potential. However EGR and ECBM are speculative options. This enhancedfossil-fuel recovery is a secondary CCS benefit. CO2 capture can increase fuel demand because of theenergy required for capture and pressurization (see Chapter 3). The fuel demand of fossil-fueledpower plants increases by 6-39%.

This chapter will explore in more depth the net effect of these different mechanisms.

Coal

The emissions per unit of energy for coal are higher than for other fuels. CO2 policies can thereforehave a negative effect on the use of coal. However, CCS can reduce the emissions per unit of coalenergy significantly, which in turn reduces the negative impacts of CO2 policies on coal use. The

167-180 chapter 7 18/11/2004 07:31

introduction of CCS results in an efficiency loss of 39% (a 12 percentage points decline of electricefficiency) for existing power generation technologies. This loss may decline to 4 percentagepoints in the long term due to the introduction of new CCS technologies in combination with newtypes of power plants (see Chapter 3). Coal demand increases due to this efficiency loss. Finally,CCS results in increased oil and gas recovery when CO2 is used for EOR, EGR or ECBM. The resultinglower oil and gas prices reduce the demand for coal. Given its perceived low importance, thiseffect has not been analysed in more detail in this study.

Figure 7.1 shows the impact of various policy incentive levels on global coal use. During the period2000-2050, coal use increases in all scenarios that include CCS. However, the introduction of policyincentives results in a decline in coal use compared to the BASE scenario. This result shows thateven if CCS is considered, coal use will decline if CO2 policies are introduced. By 2050, coal use inthe GLO100 scenario has declined by 44% compared to the BASE scenario. Note that the mostsignificant decline occurs at the lower policy-incentive levels. The additional decline of coal use fromGLO50 to GLO100 is relatively small. Note also that coal use is virtually flat up to 2030 but isfollowed by a ‘renaissance’ due to the introduction of coal-based electricity and transportation fuelcogeneration plants with CO2 capture. The results suggest that CO2 policies negatively affectcoal use, even if CCS is considered.

Fuel substitution effects that reduce coal use far outweigh additional coal use due to the lowerefficiency of power plants with CCS. This efficiency decline due to CCS is also balanced by a trendtowards higher efficiency power plants if CO2 penalties are introduced. However, the results suggestthat coal use can grow, even in a highly CO2 constrained world, if CCS is considered.

Figure 7.2 shows the decline in the use of coal if a CO2 penalty is applied but CCS is not available.The decline is expressed relative to the coal use in a scenario with the same penalty level where CCS


Figure 7.1

Coal use under various CO2 penalty levels, if CCS is considered (2000-2050)

Key point: CO2 penalties halve coal use. However, even at high penalty levels,coal use in 2050 is higher than in 2000

Co

al u

se (

EJ/y

r)

400

300

200

250

150

100

50

0

350

2010

2000

2020

2030

2040

2050

GLO50

GLO25

GLO100

GLO10

BASE

167-180 chapter 7 18/11/2004 07:31


is considered (Figure 7.1). The decline that is shown in Figure 7.2 is a measure of the impact of CCSon the future use of coal. While the benefits of CCS are limited up to 2020, they are substantialin the period 2020-2050, especially at higher penalty levels. The results show that withoutCCS, coal use would decline by more than 60% in the GLO50 scenario and by more than 70%in the GLO100 scenario. Therefore CS is of key importance for the future role of coal.

The OECD countries of Australia, the Czech Republic, Germany, Greece, Poland,Portugal, Turkey andthe US have over 65% of coal in their electricity sector fuel-supply mix. These countries would benefitsignificantly from a CO2 capture and sequestration strategy. Other countries may benefit indirectly,because the option to switch to coal with CCS as a supply substitute for natural gas and oiltransportation fuels will limit the market power of oil and gas suppliers.

Table 7.1 shows changes in coal fuel price compared to the BASE scenario in Europe and the USin five CO2 penalty scenarios, as calculated by the ETP model. The prices exclude the CO2 tax. Theimpact of a CO2 tax that corresponds with the penalty (if CCS is not applied) is indicated separately.The analysis shows that coal prices tend to increase with CO2 penalties1. Coal prices are attheir lowest in the GLO50noCCS scenario, because demand is reduced significantly. Note that aCO2 tax that corresponds to the penalty level (the last column in Table 7.1) would dwarf the fuelprice fluctuations (the centre four columns in Table 7.1). This suggests that the price impact of a

Figure 7.2

Relative change in coal use without CCS vs. with CCS under various CO2penalty levels (2000-2050)

Key point: Not deploying CCS causes a decline in coal use. The result suggests that the decline doubles between 25 and 50 USD/t CO2,

while a further increase to 100 USD/t CO2 has little impact.

Ch

ang

e in

co

al u

se (

%)

10

– 10

– 30

– 20

– 40

– 50

– 60

– 70

– 80

0

GLO100

GLO50

GLO25

GLO10

2050

2040

2030

2020

2010

2000

1. Due to upstream emissions of CO2 and methane in coal supply.

167-180 chapter 7 18/11/2004 07:31

penalty can be estimated based on the carbon content of the fuel and the penalty level, so thereis no need for more complex analysis.

Finally, the positive effects of CCS on coal gas recovery need to be mentioned. The analysis in Chapter3 suggests that about one year of current gas consumption may be recovered through enhancedcoal-bed methane recovery (ECBM). However, compared to conventional and other unconventionalgas supply options, this is a limited potential. More research is needed on the potential of ECBM.

Gas

CO2 emissions per unit of energy are much lower for gas than for oil and coal,2 although they arehigher than for renewables and nuclear energy. A CO2 policy without CCS would therefore enhancethe position of gas compared to other fossil fuels, but reduce the competitiveness of gas comparedto non-fossil fuels. It is not clear what the net effect would be on natural gas comsumption andgas prices, or whether the effect would be the same in all world regions.

CCS technology also has some secondary effects on gas supply and demand. As with coal, theintroduction of CCS will result in lower efficiency of gas use. However, the efficiency decline inrelative terms tends to be smaller for gas-fired power plants than for coal-fired power plants, dueto the higher efficiency of gas-fired power plants (see Chapter 3), so the impact on fuel use is smaller.

Figure 7.3 suggests that global gas use is not affected by CO2 policy incentives. While differencesexist in the 2010-2040 period, natural gas use doubles under all incentive levels over the wholeperiod 2000-2050. The small impact of CO2 penalties is remarkable. The result can be explainedby the fact that the coal vs. gas competition is not significantly affected by the LNG gas prices. Incertain regions coal is cheaper than LNG, while in other regions gas (either indigenous or frompipelines) is cheaper. The model results suggest almost a doubling of LNG trade, however, if theBASE scenario and the GLO100 scenario are compared.

CCS availability has a limited impact on gas use. At penalty levels up to 50 USD/t CO2, the absenceof CCS results in an increase of up to 10%. At a penalty of 100 USD/t CO2, gas use declines by up to


Table 7.1

Model coal price changes under various CO2 penalty levels, compared to BASE (2040)

Coal Additional CO2 tax

WEU (USD/GJ) (%) USA (USD/GJ) (%) (USD/GJ)

GLO10 -0.05 -3 0.03 2 0.94

GLO25 0.02 1 0.06 4 2.35

GLO50 0.14 9 0.11 7 4.70

GLO100 0.36 24 0.22 15 9.40

GLO50noCCS -0.03 -2 -0.13 -9 4.70

CO2 taxes are not included in the coal price changes.

2. Emissions of CO2 and methane in gas supply must be accounted for (upstream emissions are higher for gas than for coal).

167-180 chapter 7 18/11/2004 07:31


15%. So the impact on gas use is much smaller than the impact on coal. This pattern can be explainedby the increased substitution of coal by gas at low penalty levels. However, gas in itself is not a CO2-free energy carrier, so at 100 USD/t CO2 gas is replaced by CO2-free fuels and energy efficiency measures.The analysis in Chapter 3 also showed that the choice between coal and gas in the power sector isbarely affected if CCS is considered. If CCS is not considered, gas increases in competitiveness comparedto coal, but loses competitiveness compared to renewables and nuclear. These factors can explain therelative rigidity of gas demand under various CO2 policies with or without CCS.

The impacts of CCS on gas imports is more pronounced. By 2040, gas imports in Europe, the USand Japan are 10-15% higher in a situation without CCS. Imports in China, India and Korea areeven 40-100% higher. Clearly gas import dependency increases if CCS is not available. Also fromthis perspective, CCS increases supply security.

On the supply side, CO2 EGR accounts for 11 EJ of gas supply by 2050, equal to some 5% of total gassupply. When CCS is not considered, the development of remotely located gas and other unconventionalgas becomes more attractive from a price perspective. On a regional basis, CCS results in 10 EJ moreindigenous gas supply in the US by 2050, compared to the same scenario without CCS. In a scenariowithout CCS, this is compensated for by 10 EJ of additional gas imports from South America. The impactson other world regions are limited. In conclusion, CCS has a limited effect on global gas supply security.

Table 7.2 shows gas price changes compared to the BASE scenario in Europe and the US under fiveCO2 penalty scenarios, as calculated by the ETP model. The prices exclude the CO2 tax. The impactof a CO2 tax that corresponds with the penalty (if CCS is not applied) is indicated separately. Theresults suggest that the impact CCS has on gas price differs by region and by scenario. Gas pricestend to decline at penalty levels up to 50 USD/t CO2. However, at high CO2 penalty levels (GLO100),gas prices are considerably higher than in the BASE scenario. Higher gas prices will reduce thetendency to switch from coal to gas in order to reduce CO2 emissions.

Figure 7.3

Gas use under various CO2 penalty levels, if CCS is considered (2000-2050)

Key point: Gas use is not very sensitive to a CO2 penalty

Gas

use

(EJ

/yr)

200

40

20

0

120

80

160

180

100

60

140

2010

2000

2020

2030

2040

2050

GLO50

GLO25

GLO100

GLO10

BASE

167-180 chapter 7 18/11/2004 07:31

The impact of CCS on gas prices is not the same in Europe and the US. In Europe, prices in theGLO50noCCS scenario are lower than in the GLO50 scenario. In the US, they are higher. Note thata CO2 tax that corresponds to the penalty level would have a much more important impact thanthe fuel price fluctuations.

Oil

The production of crude and syncrude from oil sands and tar sands is shown in Figure 7.4. The ETPmodel predicts that oil production will increase by 56% between 2000-2050. The results suggestthat CO2 policies can reduce oil production and demand by between 10-20% compared to theBASE scenario, a relatively small impact. The results also suggest that these savings only takeplace at penalty levels from 50 USD/t CO2 upward.

Figure 7.5 shows the changes in oil use when CCS is not applied. At lower penalty levels there isa small increase in oil consumption of a few percent. However, in the GLO100 scenario, oil usedeclines by 10% in later decades.

CCS has a limited impact on total oil production and the regional distribution of oil supply. Themodel results suggest that Canadian production of oil sand is considerably lower in the period2020-2035 if CCS is not available. This can be attributed to the CO2 intensity of oil sand production.Oil production in the Middle East increases by 5-10% in the period 2030-2050, if CCS is not available.The share of oil production based on CO2 EOR increases to 17 EJ if CCS is considered. This represents10% of total oil production. This result depends on the modelling assumptions, however, andneeds to be studied in more detail. Given the uncertainty in the results and the relatively smallimpacts, the results do not allow for far-reaching conclusions about oil supply security benefitsthrough CCS.

In conclusion, the results suggest that CCS does not affect future oil use significantly. All scenariosshow a steady increase in oil supply. However, synfuels grow at a much faster rate than oil supply,increasing to around 100 EJ in the GLO50 scenario in 2050. This includes synthetic oil productsfrom Fischer-Tropsch synthesis, DME, methanol, ethanol and hydrogen. The steady growth ofcrude oil supply should be a topic for further analysis, given differing expert opinions as towhen a peak in conventional oil supply will occur. Most experts agree that such a peaking is


Table 7.2

Model natural gas price changes under various CO2 penalty levels, comparedto BASE (2040)

Gas Additional CO2 tax


GLO10 -0.27 -7 -0.60 -14 0.56

GLO25 -0.27 -7 -0.73 -17 1.40

GLO50 -0.05 -1 -0.66 -15 2.80

GLO100 0.60 15 0.42 10 5.60

GLO50noCCS -0.31 -8 0.74 17 2.80

CO2 taxes not included in gas price changes.

167-180 chapter 7 18/11/2004 07:31


Figure 7.4

Crude and syncrude use under various CO2 penalty levels, if CCS isconsidered (2000-2050)

Key point: Oil use is not particularly sensitive to a CO2 penalty

Oil

use

(EJ

/yr)

300

100

50

0

200

150

250

2010

2000

2020

2030

2040

2050

GLO50

GLO25

GLO100

GLO10

BASE

Figure 7.5

Relative change in crude and syncrude use without CCS vs. with CCS undervarious CO2 penalty levels, 2030 and 2050

Key point: Not using CCS causes a decline in oil use at high penalty levels

205020308

6

4

2

0

– 6

– 2

– 4

– 8

– 10

– 12Ch

ang

e in

oil

use

(%

)

GLO10

GLO50

GLO100

GLO25

167-180 chapter 7 18/11/2004 07:31

likely before 2050. However, given the importance of synfuels in the modelling results, such anearly peaking would simply accelerate the introduction of synfuels in the model. In reality, rapidexpansion of synfuel production may be a challenge.

Table 7.3 shows the changes in oil prices compared to the BASE scenario in Europe and the USunder five penalty scenarios, as calculated by the ETP model. The prices exclude the CO2 tax. Theimpact of a CO2 tax that corresponds to the penalty if CCS is not applied is indicated separately.The results suggest that both in absolute terms, and compared to the potential impact of a CO2 tax,CCS would have only a small impact on fuel prices. Impacts are similar for both regions, suggestinga small price increase. Although prices are not significantly affected by CO2 policies if CCS is available,in the GLO50 scenario without CCS they are 12% higher than in the GLO50 scenario. This can beexplained by the lack of cogeneration of electricity and transportation fuels from coal in this scenario,which increases oil demand. Note that a CO2 tax corresponding to the penalty level (the last columnin Table 7.3) would dwarf the fuel price fluctuations (the centre four columns).

Renewables

The discussion of renewables in this section is split into biomass and other renewables. The reasonfor this is because biomass represents the bulk of renewables in the GLO50 scenario (see Chapter 4).

Figure 7.6 shows the use of biomass under various CO2 penalty levels. Even the BASE scenario showsan increase of about 50% between 2000-2050. Biomass use increases with rising penaltylevels, and reaches about 125 EJ in the GLO50 scenario and 145 EJ in the GLO100 scenario,in line with the maximum biomass availability. The total amount of biomass used at higherpenalty levels is substantial and of a similar order to current global oil use. This makes biomassthe single most important renewable energy option.

Total biomass use for the residential, commercial and agricultural sectors remains roughly constantin the GLO50 scenario, compared to the BASE scenario. However, its use in industry increasessignificantly, particularly in the categories industrial boilers/process heat, industrial CHP unitsand black liquor boilers. Biomass use in the electricity sector also increases, while the share ofbiofuels used in the transportation fuel market also rises (Figure 7.7).


Table 7.3

Model oil price changes under various CO2 penalty levels, compared to BASE(2040)

Oil Additional CO2 tax


GLO10 0.37 7 0.37 7 0.73

GLO25 0.18 4 0.18 4 1.83

GLO50 0.15 3 0.15 3 3.65

GLO100 0.14 3 0.14 3 7.30

GLO50noCCS 0.64 13 0.64 13 3.65

CO2 taxes not included in oil price changes.

167-180 chapter 7 18/11/2004 07:31


Figure 7.6

Biomass use under various CO2 penalty levels, if CCS is considered (2000-2050)

Key point: If CO2 penalties are introduced, biomass use can reach current oil use levels by 2050

Bio

mas

s u

se (

EJ/y

r)

160

40

20

0

100

60

140

80

120

2010

2000

2020

2030

2040

2050

GLO50

GLO25

GLO100

GLO10

BASE

Figure 7.7

Biomass use in the GLO50 scenario (with CCS)

Key point: The potential for biomass is concentrated in industry,electricity and synfuel production

160

140

100

120

80

60

40

20

0Bio

mas

s u

se (

EJ/y

r)

2050

2045

2040

2035

2030

2025

2020

2015

2010

2005

2000

Synfuel production

Electricity sector

Black liquor boilers

Industrial CHP units

Industrial boilers/ process heat

Agriculture/ Residential/ Services heating & cooking

Note: Excludes black liquor.

167-180 chapter 7 18/11/2004 07:31

CCS has a minor effect on biomass use, as shown in the model runs with CCS (±5%). This indicatesthat the future role of biomass does not depend on the deployment of CCS.

When CO2 penalties are applied, the increase in the use of other renewables is less substantialthan for biomass in absolute terms (Figure 7.8). In relative terms, however, the increase is verysubstantial, with a fivefold increase in the GLO50 scenario and a sixfold increase in the GLO100scenario between 2000 and 2050. Most of the other renewable energy is used in the electricitysector, as described in more detail in the next section.


Figure 7.8

The use of other renewables under various CO2 penalty levels, if CCS is considered (2000-2050)

Key point: If CO2 penalties are introduced, the use of other renewablesincreases by a factor of six on 2000 levels

Oth

er r

enew

able

s u

se (

EJ/y

r)

80

20

10

0

50

30

70

40

60

2010

2000

2020

2030

2040

2050

GLO50

GLO25

GLO100

GLO10

BASE

Figure 7.9 shows the change in the use of renewables when CCS is not considered. If CCS is excluded,the use of these renewables increases by up to 40%, compared to the same penalty levels withCCS. The main increase occurs after 2020. The results indicate that CCS reduces the growth ofother renewables. However, with or without CCS, renewables will still grow rapidly.

The GLO50 scenario assumes that renewables policies are in line with the WEO 2004 ReferenceScenario (IEA, 2004a). The sensitivity analysis in Chapter 6 showed that ambitious policy targetsand technology learning effects can result in renewables increasing their share at the expense offossil fuels with CCS, which can halve electricity production from fossil fuels with CCS. The resultsregarding the impact of CCS and CO2 penalties on renewables are therefore highly dependent onthe technology learning assumptions and policy targets.

167-180 chapter 7 18/11/2004 07:31


Electricity

The availability of CCS is of greatest importance in the electricity sector. Table 7.4 shows the electricityproduction mix with and without CCS. Total electricity production is 10 EJ lower if CCS is notconsidered. This is the result of increased energy efficiency and fuel substitution in the end-usesectors. Electricity production from fossil fuels declines by 10 EJ in 2030 and by 45 EJ in 2050,resulting in more or less stable electricity production from fossil fuels. The use of coal for electricityproduction virtually disappears. Production from nuclear, hydro, geothermal and wind are considerablyhigher than in the case with CCS.

Biomass use for electricity production declines if CCS is not considered (see Table 7.4). This can beexplained by the significant co-combustion of biomass in coal-fired power plants in the GLO50scenario. This opportunity is not attractive when CCS is excluded.

CO2 policies and the availability of CCS affect electricity prices significantly. Table 7.5 shows theaverage annual electricity price increase by region. The electricity price in this analysis excludestransmission and distribution costs that would double the price for residential and commercialconsumers. Therefore relative price changes would be halved if consumer prices were compared.Since these consumer prices differ by sector, a comparison of prices excluding transmission anddistribution was chosen. The increase in the GLO50 scenario compared to the BASE scenario amounts

Figure 7.9

Relative change in other renewables use without CCS vs. with CCS undervarious CO2 penalty levels (2000-2050)

Key point: At high penalty levels, not using CCS can leadto a significant increase in the use of other renewables

Ch

ang

e in

oth

er r

enew

able

s u

se (

%)

45

35

30

25

20

15

10

5

0

40

GLO50

GLO25

GLO100

GLO10

2010

2000

2020

2030

2040

2050

167-180 chapter 7 18/11/2004 07:31

to between 5 and 73%. The additional increase when CCS is not considered amounts to between4 and 52%. Therefore the impact of not having CCS can double the impact of a 50 USD CO2

penalty. The price increases are so significant that demand may decline in the long term.


Table 7.4

Electricity production by fuel type, with and without CCS technology (2030 and 2050)

2030 2050BASE GLO50 GLO50noCCS BASE GLO50 GLO50noCCS

(EJ/yr) (EJ/yr) (EJ/yr) (EJ/yr) (EJ/yr) (EJ/yr)

FF without CCS 63.8 19.9 29.5 97.5 20.3 42.3FF with CCS 0.0 21.4 0.0 0.0 56.1 0.0Nuclear 8.7 10.5 11.9 8.4 9.6 15.4Hydro 16.2 21.0 22.3 20.2 24.1 28.0Bio/waste 5.9 12.1 11.0 8.6 15.9 12.3Geothermal 4.4 5.7 10.9 6.5 8.5 14.3Wind 1.0 10.9 12.4 6.5 18.7 28.8Tidal 0.0 0.0 0.0 0.0 0.0 0.7Solar 0.0 0.0 0.0 0.0 0.0 1.2Total 100.1 101.4 98.0 147.6 153.3 142.9

Note: FF = fossil fuels.

Table 7.5

Model electricity price increase under various CO2 penalty levels,with and without CCS technology (2040)

GLO50 Compared to BASE GLO50noCCS Compared to GLO50(%) (%)

AFR 61 35AUS 46 22CAN 33 4CHI 35 17CSA 5 52EEU 27 43FSU 70 13IND 39 34JPN 43 29MEA 71 15MEX 22 46ODA 73 18SKO 41 32USA 35 32WEU 26 6

167-180 chapter 7 18/11/2004 07:31

167-180 chapter 7 18/11/2004 07:31

Chapter 8. CHALLENGES AHEAD AND PRIORITIESFOR ACTION

H I G H L I G H T S

■ A five-fold increase in funding for RD&D on CO2 capture and storage will be needed toprepare CCS technologies for full-scale commercial introduction within 10–15 years.

■ Capture technologies: RD&D efforts should focus on innovative capture technologieswith high efficiency and low cost. Special attention should be given to the integration ofCCS into new power plant designs. At least several more projects are needed to demonstrateCO2 capture on a commercial scale. Finding sufficient funds for such projects will be asignificant challenge. Some investors may wish to proceed immediately to commercialization.

■ A new generation of highly efficient coal-fired power plants is being developed andintroduced but it will take them decades to conquer the market. This means that onlysynchronous development of a new generation of plants and CCS technologies will leadto CCS market introduction within 10-15 years. This also means that work should continueon all capture options (CCS with steam cycles, including oxy-fuelling, and CCS for gasificationcycles).

■ Storage: Sufficient proof of storage permanence is essential for any credible CCS strategyand for public awareness and acceptance. As a first step, RD&D should focus on CO2

projects which enhance fossil fuel production and on those which advance knowledge onsub-sea underground storage, and aquifer storage in locations with low population density.Stakeholder processes for reviewing, commenting and addressing concerns should be builtinto all pilot projects. Procedures for independently verifying and monitoring storage andrelated activities should also be established.

■ To facilitate the acceptance of CCS by the general public, industry decision makers, andpolicy makers, it will be necessary to make available and broadly disseminate the resultsof RD&D projects.

■ Given the controversial nature of oceanic storage, CO2 storage efforts should primarilyfocus on underground options, both off-shore and on-shore.

■ Further investment in CCS, including demonstration projects, is hindered in some countriesby uncertainties over the lack of appropriate legal and regulatory frameworks. Countriesshould create an enabling legal and regulatory environment for national CO2 storageprojects. In the interests of time, and given the diversity of institutional set-ups andregulations between countries, working at the national level using existing frameworksmay be the best short-term option.

■ Contracting parties to international instruments should be proactive in clarifying the legalstatus of CO2 storage in the marine environment, taking into consideration their objectivesto stabilize CO2 in the atmosphere.

8. CHALLENGES AHEAD AND PRIORITIES FOR ACTION 181

181-200 chapter 8 BAT 18/11/2004 07:14


■ In addition to the acceleration of RD&D funding, countries should create a level-playingfield for CCS alongside other climate change mitigation technologies. This includes ensuringthat various climate change mitigation instruments, including market-oriented tradingschemes, are adapted to include CCS.

The ETP model analysis presented in the four preceding chapters quantified the economic andenvironmental benefits of a CCS strategy. Not even the most sophisticated models can predict thefuture in detail, however. Indeed, modelling can only ever be a ‘what if’ exercise illustrating thepotential that exists. What can be concluded from the ETP analysis presented in this book is thatCCS technologies can significantly contribute to the abatement of CO2 emissions. This conclusionis valid even if, in the future, CO2 is priced very differently from the 50 USD/t CO2 assumed in theGLO50 scenario. This potential will be lost, however, if various supporting efforts to foster CCS arenot undertaken in a timely manner.

This chapter outlines the various factors associated with deploying CCS which could critically impactthe timing and effectiveness of a CCS strategy. These factors include bridging the RD&D gap, theneed for public awareness and acceptance, the importance of putting in place appropriate legaland regulatory frameworks, particularly for CO2 storage, and the need for a policy frameworkwhich encourages public-private sector co-operation and provides appropriate investment incentives.

Interrelated Challenges

There are major RD&D gaps to be bridged over the next few years if CCS technologies are to bedeveloped in time for their potential to be realised. To develop CCS technologies, significanttechnology development and deployment efforts are necessary and must be accompanied by thesimultaneous rather than the sequential development of legal, regulatory and policy frameworksand enabled by public awareness and acceptance.

At the same time, an appropriate environment must be put in place to encourage private sectorinvolvement. On the capture side, the activities by oil companies and chemical companies areencouraging. The real challenge is the introduction of widespread CO2 capture in power productionwhere the bulk of the costs arise. Investment costs for CO2 capture from a single power plant arein the order of hundreds of millions of dollars. Even for a power company which owns severalpower plants, such additional investment poses a major financing challenge. Linkage of powerplants and storage sites will imply the development of extensive CO2 pipeline ‘backbones’, towhich capture plants and storage installations can be connected. On the storage side, the bestsites and optimal storage approach need to be identified and storage permanence needs to beassured.

Most power companies do not have the resources to develop new power production technology bythemselves. That is where engineering firms come in and where government-private partnershipsmay be needed. Power producers need a clear indication that CO2 emission reductions will berewarded sufficiently over a period of decades. It is a task for government to establish these crediblelong-term policy goals and mechanisms to ensure that deep emission cuts from a single plant canbe shared by others with less promising emission reduction prospects. Governments also need toascertain that CCS really is the most economic strategy to reduce emissions and deal with energy

181-200 chapter 8 BAT 18/11/2004 07:14

security concerns. This will depend on the resource base of a country, site-specific factors, prospectsfor other emission reduction strategies, and public acceptance.

As shown in Chapter 3, different regions are in different phases of their capital stock renewalcycle. Technology development must match these cycles in order to succeed. In general, powerproducers are unable to postpone investments until better technology comes onto the market.

Finally, developing countries must be included if CCS is to be applied widely.

Timing Issues

Rapid advances in CO2 capture technology hold great promise for increasing efficiency and reducingcosts. However, power producers need reliable technology that is proven on a commercial scale. Ifappropriate investments are to be made from 2015-2020 onwards, CCS technology needs to havebeen demonstrated on a commercial scale by this date. Timing is key as the following assessmentof the likely implementation timeframe illustrates.

Planning a CCS plant could take 3-5 years, building it could take another 2-3 years and testing itan additional 3-5 years. A certain period of time may then be needed to reach full capacity andovercome operational problems. At best, such a cycle takes up to eight years. If CCS plants are tobe demonstrated by 2015-2020, this cycle must begin in the next few years. Market introductionand construction of a fourth and fifth plant would require an additional 4-5 years apiece. The bestexamples are provided by fluidized bed combustion or IGCC plants where first projects have takenup significant resources. In other words, there is very little time to start planning for the wave offull-scale CCS pilot and demonstration facilities. Therefore, the quantity and pace of work on CCSshould be increased urgently, particularly given the trends and consequences of climate change.

The four initiatives to develop megatonne-scale power plants with CO2 capture identified in Chapter 6– FutureGen in the US, the Canadian Clean Power Coalition, the Australian initiative and theEuropean HypoGen initiative – show that governments are willing to meet this challenge. However,the actual realization of these plans remains uncertain. Furthermore, four initiatives will be insufficientto establish widespread market acceptance of CCS from 2015-2020 onwards.

This tight schedule may require that less efficient technologies are combined with CCS by 2015,instead of waiting for more advanced and less costly technology. It may also require the constructionof power plants in 2015 that are suited for retrofit a decade later. This could imply building an IGCCthat would allow future low-cost CO2 capture, while a supercritical steam cycle would be a cheaperoption if CCS was not considered. The RD&D into CCS may not be completed by 2015 and acontinued effort on the scale of billions of USD may be needed over a period of decades. Whethersuch a global effort is feasible depends on the ultimate willingness to embrace the CO2 mitigationpotential that CCS offers.

It is worth mentioning that the timescale of CCS deployment depends on political and economicfactors and that there are examples of technologies going straight to large-scale use without firstpassing through the demonstration phase. Despite the additional risk of technical problems andhigher costs than if demonstration plants were built and operated first, new power plants couldbe fitted with CO2 capture technology now if incentives were high enough. CO2 storage facesmore of a time constraint because of the need to demonstrate safety and security before large-scale implementation.


181-200 chapter 8 BAT 18/11/2004 07:14


RD&D Challenges

Chapter 3 provided an assessment of the various CCS technology gaps that need to be bridged. Inparticular, it highlighted the need to reduce the cost of CO2 capture across various applications bya factor of two or more. It also highlighted the need for CCS to be demonstrated on a large scaleand for CO2 storage to be proven as feasible and environmentally safe. When such RD&D requirementsare compared with the ongoing and planned RD&D initiatives outlined in Chapter 6, it is clearthat there is quite a sizeable gap to be bridged.

With CO2 capture, governments must address the present shortage of sizeable RD&D projects inorder to advance technological understanding, increase efficiency and drive down costs. This willrequire increasing RD&D an early commercialization investment into CCS and power plant efficiency.By 2015 at least 10 major power plants fitted with capture technology need to be operating.These plants would cost between 500 million and 1 billion USD, of which half would be additionalcost for CCS.

The CCS budget is over 100 million USD per year at present. The scale of planned RD&D initiativesis too small and insufficient to ensure that CCS is implemented on a large scale in the secondquarter of this century. A five-fold increase of the funds for capture and storage projects is requiredin the short term if gigatonnes of CO2 are to be captured over the next 20-30 years. This meansthat feasibility studies for several CO2 capture and storage projects of a scale of hundreds of MWand Mt of CO2 should begin now.

The current trend for RD&D budgets runs contrary to this requirement (Figure 8.1). Governmentenergy RD&D budgets have been falling in recent decades, with total energy RD&D expenditurein 2002 just under 8 billion USD, equal to 49% of the 1980 value. The budget for fossil fuels hasdeclined from its peak of 2.7 billion USD in 1981 to 0.7 billion USD in 2002, a decline of 73%.

While the amount required for CCS is challenging, it is not insurmountable given the scale of pastenergy RD&D budgets. It would represent a 30% increase of the current RD&D budget for fossilfuels and power & storage technologies. The additional public RD&D budget that is needed forCCS development is in the range of historical R&D budgets in these categories. The proposed budgetincrease seems a challenge, but could be feasible. Leveraging the funds in private/public partnershipsis essential.

Demonstration of the safety and integrity of CO2 storage is a critical factor for technologydevelopment. A large number of data gathered through demonstration projects is needed to establishsuitable legal and regulatory frameworks, to attract financing and to gain public acceptance. Storagedemonstration projects should fully utilize early opportunities created by enhanced oil recovery(EOR) projects and sources of cheap CO2.

CO2 capture and storage have different RD&D challenges. The key issue in CO2 capture is to lowercapture costs to economically practicable levels; the processes involved are known and do notrepresent high technology risks. On the other hand, significant RD&D work is needed to prove thefeasibility and integrity of CO2 storage in various reservoirs through long-term monitoring projects.Pipeline transportation of CO2 is a proven technology and does not require significant research,and transportation by vessels is, so far, of less importance at this stage.

The cost of capturing CO2 depends on the type of power plant used, the plant’s overall efficiencyand the energy requirements of the capture process. The preferred design is for high efficiency power

181-200 chapter 8 BAT 18/11/2004 07:14

plants generating concentrated streams of CO2. The discussion below is focused on capturetechnologies in coal-based power plants which would comprise a bulk of CO2 capture facilities.As far as gas-fired power plants are concerned, R&D needs seem less challenging because theirpotential for efficiency improvements is limited as the efficiency of new plants is already high,and available post-combustion capture systems have a relatively low energy penalty. It is alsounlikely that pre-combustion capture systems based on natural gas will show a markedly superiorperformance over post-combustion capture. Novel technologies for both coal and gas, usingchemical looping and fuel cells for example, require significant basic research, meaning that theirimplementation would follow CCS deployment in power plants based on steam cycles and/or“regular” IGCC.

Increasing the efficiency of fossil-fuel power plants is a powerful CO2 abatement measure on itsown. Several roadmaps have been proposed which include two interconnected paths of technologydevelopments. The first trajectory leads to increasing efficiency of power plants, and the secondone to including CCS in power systems. Both trajectories eventually merge (Figure 8.2). The timingof the merger depends on RD&D developments as well as on a process of monetisation of CO2 abatement, the introduction of legal and regulatory frameworks and on levels of publicacceptance.

The two diagrams below (Figures 8.3 and 8.4) outline roadmaps for efficiency improvements andCCS development in OECD and non-OECD countries as proposed by the IEA Clean Coal Centre(Henderson, 2003). With both diagrams, the first path represents goals for pulverised coal steamcycles, the second path concerns IGCC plants, and the third the implementation of CO2 capture(efficiencies given in Figures 8.3 and 8.4 differ slightly from the data provided in Chapter 3).


Figure 8.1

IEA government RD&D budgets

Key point: Budgets have been declining over the lasttwo decades and currently amount to 8 billion USD per year

1974

1976

1978

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

18000

8000

2000

16000

12000

4000

10000

14000

6000

0Mill

ion

US$

(2

00

2 p

rice

s an

d e

xch

ang

e ra

tes

Other

Nuclear Fusion

Power & Storage techologies

Nuclear Fission

Fossil Fuels

Conservation

Renewable Energy

181-200 chapter 8 BAT 18/11/2004 07:14


As Figures 8.3 and 8.4 illustrate, coal-fired power plants based on steam cycles and IGCC technologyhold great promise for further efficiency gains. RD&D into steam cycles is focused on developing ferrousalloys and nickel-based super-alloys for higher steam conditions, further improvements in steam turbineand the introduction of oxy-coal combustion. RD&D on CO2 capture from this type of plant is focusedon developing new chemical and physical solvents for CO2 scrubbing, membrane and adsorptionseparation techniques and, in general, minimizing the energy required for CO2 capture.

Work on improving IGCC plants includes improving refractories, gas coolers and coal feeding systemsto increase reliability and the availability of installations, improving coal conversion and co-gasification, dry gas clean-up, turbines for synthesis gas and hydrogen, and air separation for O2

production. Improvements are also being made to fuel cells to scale up and demonstrate their useof hydrogen from synthesis gas. RD&D for CO2 capture involves developing CO2 separationtechnologies for fuel gas.

The technology developments and efficiency improvements outlined above cannot be taken forgranted, however. Major efforts need to be undertaken and significant resources committed bygovernments and industry to realize these projects. The COORETEC programme, recently launchedby the German government in co-operation with industry, is an example of an ambitious initiativefocused on efficiency improvements with parallel projects on CCS (COORETEC, 2003).

So far, the significant work done on CO2 storage suggests that CO2 can be stored for thousandsof years or longer if held in a suitable reservoir. Storage risks are known and have been shown tobe low, providing the right facility is chosen. Furthermore, RD&D has developed appropriatemonitoring techniques that can be applied at reasonable cost, as well as remediation techniques.Nonetheless, further RD&D is required, particularly into the following areas (Benson, 2004):

● investigation of storage effectiveness through monitoring CO2 storage sites, studying analogues,basic research on physicochemical processes involved in trapping CO2 and numerical simulation;

● estimating and proving storage capacity on a global, regional and national level;

Figure 8.2

Trajectories for increased efficiency and CCS development

Near-term Medium-term Long-term

CO2 emission reduction

GLO50

Increased Efficiency Trajectory

ZeroEmissionsTrajectory

Source: Otter, 2004.

181-200 chapter 8 BAT 18/11/2004 07:14

● environmental risk assessment, identification of risks involved and the development of appropriatesite selection procedures, monitoring techniques and remediation actions, and the developmentof a regulatory framework;

● demonstration projects to further prove the viability of long-term storage.

Figure 8.5 shows the cumulative capacity of major ongoing and planned storage demonstrationand monitoring projects. In order to ensure the projected exponential growth, a far larger numberof such projects are needed to fully validate the CO2 storage concept. The general public, includingnon-governmental organisations (NGOs), must be involved in CCS development at every stage.Storage is a key area where gaining public acceptance is of critical importance. This includesacceptance for a CCS strategy in general and also local acceptance of specific storage projects.

Public-private partnerships have a crucial role to play in financing RD&D activities. To encouragetheir involvement, governments must take into consideration the various objectives and prioritiesof different stakeholders. Oil and gas companies will need CO2 removal for gas purification. CO2

EOR is likely to be an economically attractive option for many reservoirs. The expertise oil and gascompanies offer for gas injection techniques and reservoir management would be useful for allCCS geological storage projects, not only those involving gas and oil fields. Manufacturing companiesworking for the oil and gas industry are likely to be interested in CCS for similar reasons; coalcompanies would be interested in CCS to ensure a market for their product, while electricity utilitiesmay be interested in CCS to help them prepare for the introduction of CO2 abatement fiscalmechanisms. In general, corporate responsibility could be a powerful driving force for many companies.


Figure 8.3

Roadmap for efficiency improvements and CCS development in coal-firedpower plants in OECD countries

Clean coal technologies

Path to near-zero emissions

SCSC 40-45 % USCSC 50-55 % USCSC 50-55 %SCSC 47 %USCSC demo 50 %

Efficiency improvements;

better environmental control

Retrofit SCSC with chemical absorption

30 %

SCSC + CA 35 %

Retrofit IGCC + PA 35 %

SCSC Oxyfueling

40 %IGCC PA

40 %

IGCC membrane techs 45 % IGCC-SOFC

55-60 % IGCC for

synfuel cogeneration

IGCC 50 %IGCC-SOFC demo

Increasing efficiency, lower emissions, lower costs

Now 2020

IGCC 47 %IGCC 50 % +

(IGCC-SOFC below)IGCC demo 45 %

Source: Henderson, 2003.

181-200 chapter 8 BAT 18/11/2004 07:14


Figure 8.4

Roadmap for efficiency improvements and CCS development in coal-firedpower plants in non-OECD countries

Clean coal technologies

Path to near-zero emissions

USCSC 50 % USCSC 50 % +Subcrit. SC 40-45 %

Efficiency improvements;

better environmental control

SCSC 45 %

SCSC CA 35 %IGCC

45-50 %

IGCC PA 40 %

Increasing efficiency, lower emissions, lower costs

Now 2020

IGCC 40-45 % IGCC 50 % +

SCSC 45 %

Mt

CO

2

1

10

100

1,000

10,000

100,000

1,000,000

Sleipner

Weyburn

In Salah

Snohvit

1990

2000

2010

2020

2030

2040

2050

GLO50

OECD50

Gorgon

Source: Henderson, 2003.

Figure 8.5

Major CO2 storage projects and the uncertain long-term developments(cumulative)

181-200 chapter 8 BAT 18/11/2004 07:14

Public Awareness and Acceptance

The deployment of CCs technologies will require broad understanding and long-term commitmentby numerous constituencies (McKee, 2003). These include, among others, central and localgovernments, the general public, environmental and non-environmental NGOs, industrial andcommercial organisations, academic and scientific institutes, financial institutions, the media, andinternational organisations. The following discussion on awareness and acceptance is limited tothe general public and environmental NGOs.

Until recently, CCS technologies were of interest to a relatively small group of experts. This meansthat, at present, public awareness of CCS is very limited. A recent review (Curry and Herzog, 2004)shows that few people know anything about CO2 capture and storage or understand the relationbetween CCS and climate change. The survey also revealed a poor understanding of CO2 sources,mechanisms driving climate change and mitigation measures.

Awareness of the potential of CCS is the first step towards gaining acceptance for its deployment.If CCS is to be widely accepted, a policy of ‘openness’ is required. All communication efforts shouldbe based on high quality data. National consultation and regional negotiations are critical to thesuccess of CCS projects since, by its very nature, the technology would require large industrial-sizedprojects affecting local and regional communities.

Two types of opposition can be discerned at this stage: opposition from stakeholders who preferother mitigation measures to CCS (or object to the use of fossil fuels and CCS completely), andlocal opposition to specific projects, notably storage. Independent credible analysis, based onscientific data which addresses the real risks and also pros and cons of CCS and associated projects,is required for any ultimate acceptance.

To facilitate acceptance of CCS by the general public, industry decision-makers, and governmentpolicy makers, it will be necessary to develop well-structured education and outreach programmes(Esposito and Locke, 2003). The absence of organised, effective communication strategies, controversyand fear of leakage could pose an obstacle to scientific research and CCS deployment. Such problemshave already arisen for two oceanic storage R&D projects (in Hawaii and Norway). With coal-bedmethane recovery in the US, locals have suffered from a lowering of groundwater levels, a deteriorationin surface water quality, and soil pollution, among other things. As a result, considerable oppositionhas built up towards coal gas extraction in general (Powder River Basin, 2004). Similar problemscould arise for ECBM projects. The lesson that can be drawn from these experiences is that involvingstakeholders at an early stage is essential to mitigating major development problems. Given thedispersed nature of potential CO2 storage sites, development should focus on areas which are lessecologically sensitive than others, even if this incurs additional costs. Stakeholder processes forreviewing, commenting and addressing concerns should be built into all pilot projects, togetherwith procedures for independently verifying and monitoring storage and related activities.

Environmental NGOs understand the need for a deep cut in GHG emissions and the majority ofNGOs consider CCS to be a potential bridging technology on the way to a CO2-free energy systembased on renewables. NGOs generally support RD&D work on CCS technologies (Goerne, 2004).Their main concern centres on the fact that CCS is seen and presented as a solution which wouldallow for the continued use of fossil-fuel resources as long as they are available.


181-200 chapter 8 BAT 18/11/2004 07:14


NGOs are also concerned about the following factors, among others (Craig 2003):

● CCS may divert resources away from alternative emission mitigation options such as increasedrenewable energy use and energy efficiency.

● CCS may give false hope to those who could regard it as the ‘silver bullet’ of CO2 mitigation.This could set back other climate change policies in the short and medium term.

● CCS leads to additional energy use.

● Environmental issues associated with the impact of fossil-fuel extraction and transportationremain.

● The risk of CO2 leaking from storage sites.

● CCS results in a 40-80% increase in the cost of electricity.

● The competitiveness of CCS in relation to renewables and energy efficiency measures stillneeds to be established.

These issues have been analysed in the previous chapters. NGOs strongly object to CO2 storage inthe oceans (in the water column) because of its potentially harmful impact on the marine environmentand the fact that CO2 could diffuse to the ocean surface and eventually reach the atmosphere. Alack of scientific data and uncertainty over the behaviour of CO2 does not allow for any larger-scale pilot project. Thus, injection into the water column is not being widely pursued as a viablestorage option for the time being. This does not present a significant problem, however, as otherstorage options based on geological storage represent sufficient capacity on a global scale. Thatsaid, serious academic research on ocean storage is being undertaken in some countries, notablyJapan.

The Regulatory and Legal Framework

National and international legal and regulatory frameworks for CCS need to reflect scientific andtechnological progress as well as the various objectives of the stakeholders and the internationalcommunity. The legal and regulatory frameworks currently applied to CCS were established beforeit emerged as a viable technology and environmental policy option, before climate change mitigationbecame a priority among the international community. These frameworks will need to be updatedto take into account the scientific progress that has been achieved in CCS and in light of the newgreenhouse gas mitigation objectives.

On-shore storage primarily falls within the scope of national legal frameworks. CO2 storagedemonstration projects, including EOR with CO2 storage, are being carried out in several countriesunder a myriad of non-CCS-specific regulations, such as those governing oil and gas activities, mining,pipelines, transport, environmental impact assessment, property or liability.

Off-shore storage primarily falls under the international legal framework governing the marineenvironment. Under this framework, large-scale offshore projects will face legal uncertainties existingunder the London Protocol and the OSPAR Convention.

Cross-border exchanges of CO2 and storage of CO2 in or under international waters may pose anumber of specific liability issues in the context of the UN Framework Convention on Climate Change(Haefeli et al., 2004). National and international legal and regulatory frameworks are discussedin more depth in the publication Legal Aspects of Storing Carbon Dioxide (IEA, 2004c). Major issuesand priority actions are summarized below.

181-200 chapter 8 BAT 18/11/2004 07:14

National regulatory frameworks for onshore CCS activitiesLegal and regulatory situations vary considerably from one country to another depending on thefossil fuel resources available, what stage each country is at with CO2 storage technologies, andspecific public acceptance concerns. Countries with mature oil and gas reserves tend to have moreexperience with CO2 storage through CO2 EOR and acid gas injection than others.

Each of the various activities involved is governed by existing laws, such as those covering oil andgas activities, mining, pipelines, transport, environmental impact assessment, property or liability.Therefore, CCS activities potentially fall within the scope of many regulations. Carrying out acomprehensive due diligence of the applicable framework is an expensive exercise. In general,existing frameworks are better suited to the capture and transport stages of CCS than to injectionand storage.

Regulatory gaps are associated with long-term storage, site characterization, monitoring and liability.Countries declare a lack of empirical understanding of associated risks to fully assess these gapsand thus improve their regulatory framework. The other main gap is the inclusion of CCS in climatechange mitigation mechanisms.

United States

In the US, there are two levels of legal and regulatory framework in accordance with the allocationof powers between the Federal and State governments. At the Federal level, the EnvironmentalProtection Agency currently considers that CO2, like other greenhouse gas emissions, is not an airpollutant subject to regulation under the Federal Clean Air Act1. There are no Federal laws explicitlygoverning each stage of CCS, namely capture, transport, injection and post-injection.

There is, however, a large body of existing Federal law governing interstate pipeline activities,hazardous wastes and underground injection wells and their controls. These could be adapted toencompass CO2 storage activities. Furthermore, there is a large body of existing Federal case lawdistinguishing between EOR, storage and waste disposal for the purposes of classifying injectionactivities. Whether or not the substance being injected has a commercial value is an importantcriterion for determining whether it is categorized as a waste when being stored. This might havea bearing on the determination of any future framework for CO2 storage.

At the State level, there are a significant number of regulations governing CO2 capture, transportand injection, developed for the oil and gas industries. Site ownership issues also fall under thejurisdiction of State law, which may vary considerably from one State to another. Given thisinstitutional structure, regulating CCS in the US will not be a ‘one stop shop’. Some powers mightbe vested with the Federal government, but most will be vested with the State.

Whichever mix is eventually chosen, there is already a substantial body of Federal and State lawthat could be adapted to encompass CCS activities and thinking on how to apply it to CCS. Whetherreform will come from individual States or the Federal government will depend largely on howexisting Federal laws are interpreted, including the Clean Air Act. Should it be decided that Federallaws do not apply, there will be more room for States to step in.


1. At the time of writing, this issue is before the Federal courts, where environmental NGOs are suing the EPA on the grounds that CO2

is an air pollutant.

181-200 chapter 8 BAT 18/11/2004 07:14


United Kingdom

There is a large body of regulations applicable to onshore CCS activities in the UK,2 althoughthese were generally not designed with CCS activities in mind.

Regulations applicable or potentially applicable to onshore storage of CO2 include the PetroleumAct, the Pollution Control Act, the Planning and Building Act, the Chemical Regulations, legislationcovering dangerous goods, health and safety legislation, Regulations to the Petroleum Act, andthe Major Accident Hazards Regulations. In addition, all CO2 storage activities would have to complywith applicable EU regulations, including the Contaminated Land and Health and Safety Directiveand the Water Framework Directive. There is no existing case law on CO2 storage in the UK, butthere are precedents on gas storage.

Overall, the existing framework is not likely to prohibit CO2 storage. Adapting it to take into accountcapture and transport activities is not expected to raise particular problems. Injection and storageactivities, on the other hand, could lead to issues which would need to be addressed.

According to a study carried out for the British government, there seems little doubt that CO2 wouldbe classified as waste if permanently stored (‘disposed of’) because CO2 has no value and, therefore,there would be no intention to recover it at a later stage. For CO2 EOR or ECBMR, the classificationof CO2 could depend on the value placed on the delivered CO2. If CO2 is considered waste, itsstorage would be governed by applicable EU regulations as transposed in UK law, such as the WasteFramework Directive and the Landfill Directive.

The most important legislative gaps concern the status of CCS within the market-based and regulatoryframework to address CO2 abatement, in particular the emission trading scheme, and the long-term monitoring and ownership issues associated with it. Emissions data from offshore injectionwould have to be provided to the UK Greenhouse Gas inventory.

Japan

There is no legal or regulatory framework explicitly applicable to CO2 storage in Japan. As of July2004, there was only one CCS field experiment being conducted in Japan. The research instituteresponsible for this is acting under the existing legal framework – mainly the Road Traffic Law, theHigh Pressure Gas Safety Law, the Mining Law, the Mining Safety Law, the Agricultural Land Law,the Water Control Pollution Law and the Waste Disposal Law. All responsibilities for the project liewith the research institute. This project is conducted under existing laws because it is experimentaland small in size. Additional regulation would have to be drafted for larger projects.

Canada

Like the US, the Federal government and the Provinces of Canada have different jurisdictions overCCS activities. Resource ownership and development fall under the jurisdiction of the Provincialgovernments. The Federal Government has jurisdiction when trans-boundary or trade andenvironmental issues are involved.

There are currently two ongoing CO2 EOR projects in Canada and almost fifty acid gas (H2S) injectionschemes for disposal and containment. Four additional demonstration projects may be coming upin Alberta in the coming years. Although there are no particular incentives in the market to encourage

2. Given its oil and gas resources, the UK is more interested in offshore storage, which would be governed by international frameworksand regulations specifically applicable to offshore activities under the jurisdiction of the Crown.

181-200 chapter 8 BAT 18/11/2004 07:14

private operators to engage in long-term storage, the Canadian government is strongly encouragingany CO2 EOR initiative as well as any longer-term storage and monitoring initiatives. A federalCO2 capture and storage incentive programme and an Alberta royalty credit programme havebeen initiated to further stimulate commercial demonstration projects in CO2 –based resourcerecovery.

Federal and Provincial frameworks that may apply include legislation governing land administration,land-lease, explosives and dangerous goods, petroleum safety, pipelines, mineral resourcesdevelopment, occupational health and safety, planning, coal mining safety, the environment andoff-shore activities. None of these frameworks was specifically designed to address CCS.

Like other countries, existing legal frameworks in Canada adequately cover or could be modifiedto cover the capture, transport and possibly injection stages of CO2. There are serious gaps, however,regarding long-term storage issues such as monitoring and liability. In addition, there is no frameworkgoverning the valuation of CO2 stored, emission reductions and emission permits.

The development of relevant frameworks in Canada is likely to follow scientific progress and knowledgeacquired from the various CO2 storage projects conducted in Canada.

Australia

The Federal Government and the States of Australia have different jurisdictions over CCS activities.

There is no legal and regulatory framework specific to CCS activities in Australia, except for oneproject-specific legislation for the Gorgon Project in Western Australia. Applicable legislation includeslaws governing occupational health and safety, the environment, petroleum activities, mineralresources, dangerous goods, coal mining safety and health, offshore activities, land lease, landadministration, explosives and dangerous goods, pipeline and planning. In addition, offshore geo-sequestration might be considered as dumping under the Dumping Act.

Australia recognizes the existence of legal and regulatory gaps for CCS. Accordingly, it has beenagreed that the Federal and State governments will work together to develop a common andconsistent national framework to cover all aspects of CCS regulation in the country.

The approach taken has been to prepare a draft set of non-binding regulatory principles which willbe submitted to a ministerial council for endorsement. Each individual jurisdiction would then decidewhether, when and how to implement them.

Because many of the issues involved with CCS are already covered by existing legislation, it isexpected that implementing these principles would mostly be accomplished by amending existinglegislation rather than drafting new laws.

Access and property rights as well as long-term liabilities are considered to be the issues on whichmost work still needs to be done. Community consultations to raise community awareness areconsidered paramount and have started in some areas.

The EU

There are several EU directives that are potentially applicable to CCS: namely, the Framework Directiveon Waste Materials (75/442), the Directive on Dumping of Waste Materials (1999/31), theEnvironmental Impact Assessments Directive (85/337 as amended by Directive 97/11) and the


181-200 chapter 8 BAT 18/11/2004 07:14


Framework Directive on Water (2000/60/EG)3. These Directives were not designed with CCS inmind and, at the time of writing, no CCS legislation is being drafted in Brussels.

The applicability of the Directives to CCS will, therefore, be determined separately by each EUmember state, on the basis of their various implementation instruments. Also relevant in the EuropeanUnion is the EU Emissions Trading Scheme (ETS) which allows CCS subject to the establishment ofsatisfactory monitoring and reporting guidelines.

International legal frameworks for offshore storage The main international legal frameworks relevant for CO2 storage are those governing marineenvironment protection and climate change. These embody two of the main environmental objectivesof the international community – stabilizing the atmosphere and protecting the hydrosphere andits environment – which have so far been pursued independently from one another despite sometimeshaving overlapping scopes.

The marine protection framework, which was established before the emergence of CCS as a majorCO2 emissions reduction option, contains significant constraints on offshore CO2 storage activities.By contrast, the climate change framework has yet to deliver effective CO2 emission reductionsobligations on contracting parties and incentives for CCS development.

How to combine the respective objectives of these frameworks in the face of technological changeand growing knowledge of climate change is one of the main challenges to the development ofan enabling international legal framework for CCS. The main international conventions and theirstatus are listed below (Table 8.1).

Marine ProtectionInternational marine environment protection was established in 1972 with the London Conventionto regulate the dumping of wastes and other matter at sea. In 1982, this field was extended throughthe adoption of the United Nations Convention on the Law of the Sea (UNCLOS). As an overarching

3. The Framework Directive on Water aims to ‘maintain and improve the aquatic environment in the Community.’ The Directive definesa pollutant as ‘the direct or indirect introduction, as a result of human activity, of substances or heat into the air, water or land whichmay be harmful to human health or the quality of aquatic ecosystems or terrestrial ecosystems directly depending on aquatic ecosystemswhich result in damage to material property, or which impair or interfere with amenities and other legitimate uses of the environment.’CO2 is not on the Directive’s lists of pollutants or dangerous substances, but potential points of contention include whether CO2 injectionand storage could affect ground and surface waters.

Table 8.1

Main international conventions relevant to CCS

Convention Subject Signature Entered into force

UNCLOS “Constitution” of the seas 1982 Yes

London Convention Marine protection 1972 Yes

London Protocol Marine protection 1996 No

OSPAR Convention Marine protection 1992 Yes

UNFCCC Climate change 1992 Yes

Kyoto Protocol Climate change 1997 No

181-200 chapter 8 BAT 18/11/2004 07:14

agreement, UNCLOS does not contain detailed operative provisions on most maritime issues; rather,it provides a framework for all areas, including marine protection, and allows other, more targetedtreaties to fill in the gaps.

With regard to marine pollution, the standards are set by the Convention on the Prevention ofMarine Pollution by Dumping of Wastes and other Matter, signed in London in 1972 and knownas the London Convention. Underneath the London Convention fall several regional agreementscovering specific areas of the ocean. The most widely known of these is OSPAR, the Convention for


UNCLOS and the legal zones of the sea

The conditions under which the various international maritime agreements apply to CO2storage depend on the location of storage sites within one or other of the specific legal zonesof the sea defined by UNCLOS: the Territorial Sea, the Exclusive Economic Zone (EEZ), andthe High Seas (Figure 8.6). A country’s territorial sea constitutes the band of ocean stretchingup to 12 miles from its shores. Within this area, nations’ ‘sovereignty over the TerritorialSea is exercised subject to ... rules of international law.’

A nation’s EEZ extends from the end of the Territorial Sea out to 200 miles from a country’scoast (i.e., 188 miles from the end of the Territorial Sea). Coastal states have sovereign rightsto explore and exploit the natural resources of the seabed and subsoil of the continentalshelf [land which is usually contained within the EEZ].’ Beyond this area are the High Seaswhich are open to all countries. However, each country is entitled to complain if the activitiesof others cause undue harm to their interests.

Continental shelf

Continental slope

Continental rise

12 miles

Mainland

High seas

Terri

toria

l sea

Cont

iguo

us Z

one

200 milesExclusive economic zone

Deep seabed

12 miles

Figure 8.6

UNCLOS legal zones of the sea

Source: IEA GHG R&D Programme, 1996

181-200 chapter 8 BAT 18/11/2004 07:14


the Protection of the Marine Environment of the North-East Atlantic. OSPAR’s regulations on marinepollution are markedly stricter than those of the London Convention, and, unusually, its decisionsare legally as opposed to politically binding on its Contracting Parties.

The relevance of the London Convention to CO2 storage is limited as it only applies to storageconducted from vessels, platforms and other man-made structures in the water column. Consequently,it does not apply to storage in saline aquifers or any other geologic formations. In addition, theLondon Convention only prohibits CO2 storage in the water column if CO2 is considered as industrialwaste, which is still an open debate requiring clarification. Some discussions on CO2 storage wereheld within the London Convention in recent years, including whether CO2 should be classed anindustrial waste. No definitive conclusions were drawn, however. The Scientific Group establishedunder the London Convention has a watching brief on the issue.

The London Convention also requires Contracting Parties to be guided by the precautionary approachto environmental protection when implementing their obligations under the Convention. Accordingto this principle, appropriate preventive measures must be taken when there is reason to believethat substances or energy introduced into the marine environment are likely to cause harm evenwhen there is no conclusive evidence to prove a causal relationship between input and effect.4 Ithas been argued that this principle would prevent ocean storage of CO2 even if CO2 is not consideredan industrial waste. However, it has also been claimed that it is not yet clear whether storage withimpermeable caps would be considered as a risk to the marine environment. No definitive legalposition has been adopted on this issue, whether by the Consultative Meeting of Contracting Parties,the International Court of Justice or other international entity with jurisdiction over the matter.

The London Protocol has not been ratified yet. However, its remit is far wider with regard to dumpingat sea than the London Convention. The dumping that applies to both comprises:

● deliberate disposal at sea (including in the water, seabed and subsoil but not territorial watersof states) of wastes loaded on board a vessel and from offshore installations; and

● any storage of wastes in the seabed and the subsoil.

In addition, the London Protocol circumvents the waste definition issue by prohibiting all dumpingexcept for wastes listed on a ‘reverse list’, of which CO2 is not a part. However, dumping under theProtocol does not include pipeline discharges from land, operational discharges from vessels oroffshore installations or placement for a purpose other than disposal, if such activities do not runcontrary to the aims of the protocol.5 Subject to these exceptions, the London Protocol would,therefore, prohibit without distinction the storage of CO2 both in the water column and in the sub-seabed.

The OSPAR Convention, established in 1992 by 15 north European member states and the EuropeanUnion,6 is considered by far the most comprehensive and strict legal framework governing the marineenvironment. Although not drafted specifically with CO2 storage in mind, some of its provisionsare interpreted as creating significant constraints on any offshore CO2 storage activities. The OSPARcommission is developing an agreed position on whether placing CO2 in the sea and the aquifersbelow the sea is consistent with the OSPAR Convention. This highlights the legal uncertaintyconfronting potential offshore investors.

4. Resolution LDC.44(14), 1991.5. Whether CO2 storage may constitute such a placement is still open to question.6. It is also used as a guideline for marine environment protection by non-OSPAR contracting parties.

181-200 chapter 8 BAT 18/11/2004 07:14

Atmosphere stabilization

The climate change framework was established in the early 1990s to restrain man-made emissionsof greenhouse gases. It consists primarily of the UN Framework Convention on Climate Change(UNFCCC) signed in 1992 and effective since 1994, the Kyoto Protocol adopted in 1997 and regionaland national implementing instruments.

The main objective of the climate change framework is to stabilize the concentration of greenhousegases, including CO2, in the atmosphere by reducing emissions. The UNFCCC does not create bindingobligations upon countries to reduce CO2 emissions but promotes, in general terms, the utilizationof carbon sinks. The Kyoto Protocol creates binding obligations on a minimum number of developedcountries to reduce their CO2 emissions by 5.2% below 1990 levels through a system of emissionquotas and emission trading. The entry into force of the Protocol is likely, given the recent Russiansignature.7

Neither the UNFCCC nor the Protocol expressly include or exclude CCS as an encouraged or permittedemission reduction device giving rise to emission credits. Should the Kyoto Protocol enter intoforce, the status of CCS would have to be clarified in order for it to reap the benefits provided bythe Protocol, in particular those of emissions trading. This includes establishing whether or notsignatory countries could account for CCS in national inventories (Haefeli et al., 2004).

Priority Actions

To overcome legal uncertainties for investors, the following priority actions are recommended.

Additional storage and monitoring projects need to be carried out to fully assess long-term storagerisks and establish purposeful and consistent siting and monitoring requirements. Ongoing EORprojects have not focused on long-term storage aspects and there are too few storage projects withdetailed monitoring components to be of large-scale use. Empirical data and close co-operationbetween the scientific community, industry and regulators will be essential to establish standardsfor regulatory and legal frameworks and address public acceptance issues for CCS.

In the short-term, governments should provide the appropriate national legal environment forincreasing the number of storage demonstration projects. Longer term, national frameworks shouldbe formulated on the basis of adequate empirical knowledge of the conditions and risks of long-term storage.

Contracting parties to international instruments should take a proactive stance in clarifying the legalstatus of CO2 storage in marine environment protection instruments, taking into consideration notonly their marine environment protection objectives, but also those regarding climate change mitigation.Similarly, clarification of issues relating to cross-border movements of CO2 might be needed.

Long-term Policy Framework and Incentives

As outlined earlier in the chapter, public-private funding is needed for RD&D to get to the marketdeployment stage. In addition, the full-scale commercial deployment of CCS will require appropriate


7. The Kyoto Protocol will enter into force after at least 55 Parties to the Convention - incorporating Parties which accounted in total forat least 55% of the total CO2 emissions for 1990 from the group of industrialized countries - have ratified it and completed all formalities.

181-200 chapter 8 BAT 18/11/2004 07:14


remuneration of investors for the additional capital and operating costs of CO2 control installations.In order to support development and deployment of CCS, companies need a clear indication thatCO2 emission reduction will be rewarded sufficiently over a period of decades. It is the responsibilityof governments to establish credible, long-term policy frameworks and incentives. This chapteronly briefly mentions CO2 mitigation policies and incentives under discussion. This issue will bethe subject of a follow-up publication.

Investors are expecting long-term certainty about investment and fiscal incentives and/or a CO2

pricing mechanism. This latter option could be in the form of a fixed CO2 tax or use of a flexiblemechanism based on a market response to abatement policies. When introduced, flexible mechanismswould initially set the CO2 price at a level adequate to cover costs of the cheapest abatementoptions.8

CCS is unique in the sense that it is applied to large point sources where it results in deep emissioncuts. A regulatory approach aiming for CO2 emission reductions of a few percent from each powerproducer is not an appropriate tool with which to kick-start CCS. Flexible mechanisms will be neededwhere the credits can be traded, or where a carbon tax could be used.

Carbon taxes are viewed favourably in some countries, although debate on their effectivenesscontinues in others. Countries which have introduced carbon taxes include Finland, Sweden, Germany,the UK, the Netherlands and Norway. Norway’s carbon tax has been instrumental in fostering theSleipner project – covering the cost of CO2 pressurization and storage. Carbon taxes are also beingactively considered in many other parts of the world. In most cases, it is expected that only largeCO2 emitters would be targeted.

Under the terms of the Kyoto Protocol, three flexible mechanisms are defined:

● The Clean Development Mechanism (CDM). This was created to allow industrialized nations tomeet part of their emission reduction targets by cutting emissions in developing countries,providing this contributes to the sustainable development of the host country. The CDM isexpected to result in increased investment into developing countries, thereby fosteringenvironmentally beneficial projects that might not otherwise have been feasible.

● Joint Implementation (JI). This constitutes the other project-based mechanism defined in the KyotoProtocol to allow the joint implementation of greenhouse gas reduction activities within countrieswith agreed reduction targets (38 industrialized countries, including 11 in Central and EasternEurope). This enables participating countries to work together to meet their respective targets.

● Greenhouse Gas Emission Trading Schemes (ETS). These are briefly reviewed below.

All of the above will help to encourage greater international co-operation, leading to more rapid andeffective development and application of CO2 control strategies. It is still unclear, however, howpolicy makers will consider the eligibility of CCS for the CDM and JI mechanisms – and it is not clearhow CCS activities should be accounted for in national inventories to demonstrate compliance withthe Kyoto objectives. There is a prevailing opinion that no special agreement should be sought concerningthe eligibility of CCS, rather that ‘testing the water’ projects should be packaged and submittedaccording to general requirements. Questions relating to ways of accounting for CCS activities and

8. CO2 abatement cost curves for deploying renewables and energy efficiency measures indicate that there are still many cheaper optionsthan CCS. Their potential, however, is limited. After a certain point, they would increase in cost with a CO2 price gradually rising to levelsadequate for implementing CCS. The process may take years and would take longer in developing countries than in developed economies.It is estimated that, within the EU, implementing options cheaper than CCS would be a viable strategy only for the next 10-15 years.

181-200 chapter 8 BAT 18/11/2004 07:14

establishing appropriate baselines for CDM or JI projects based on CCS technologies have beeninvestigated by the IEA’s Energy Efficiency and Environment Division (Haefeli et al., 2004).

Greenhouse gas emissions trading schemesA greenhouse gas emissions trading scheme is a market-based mechanism that allows emissionreductions achieved by one party to be sold or passed on to a second party. It is generally assumedthat the assigned emissions target for the country in question will ultimately be passed on toindividual enterprises and commercial organisations in the form of emissions caps. Those withemissions below the cap will be able to sell excess credits to another party; levels in excess of thecap will require the purchase of additional credits from elsewhere. This concept is not new; similarschemes for SO2 trading have been operating successfully in some parts of the world for a numberof years. Extensive modelling studies carried out suggest that adopting a comprehensive carbontrading scheme will be instrumental in cutting national CO2 emissions.

Greenhouse gas trading schemes have a role to play in encouraging further development and useof carbon control strategies. When a commercial or industrial enterprise adopts appropriate CO2

control measures, this creates the potential for bringing emission levels down to below the agreedcap value. If such a potential is realised, excess credits can be traded or sold, generating an additionalsource of income.

An EU Directive for emissions trading comes into effect in January 2005. During the first phase,(2005-2007), CO2 will be tackled through a ‘cap and trade’ system, concentrating initially onemissions from large industrial and power generation activities. It is estimated that the schemewill affect roughly 45% of total EU emissions of CO2 projected for 2010. Organisations that failto meet their agreed targets will be required to pay a harmonized penalty charge, while those withexcess credits will be able to trade these with third parties. Several EU member states already havetheir own schemes in place and discussions are in hand on how to harmonize these. So far,12,000 installations in Europe have been ‘capped’ with opening trading prices estimated to bearound 8-15 USD (7-13 EUR).

Although it is not specifically mentioned, CCS is likely to be eligible for trading under theDirective but would require the establishment of national storage monitoring and verificationguidelines by each country. Renewables and other zero emissions energy technologies are supportedby independent incentive schemes.

Countries should create a level-playing field for CCS alongside other climate change mitigationtechnologies. This includes ensuring various climate change mitigation instruments, including market-oriented trading schemes, are adapted to include CCS. The future role of CCS depends critically onsufficiently ambitious CO2 policies in non-OECD countries. Therefore, outreach programmes todeveloping countries and transition economies and international commitment to reduce CO2

emissions is a prerequisite. The maturation of a global emissions trading scheme, a meaningfulprice for CO2 and a predictable return on investment are important factors that could stimulatethe timely deployment of CCS.


181-200 chapter 8 BAT 18/11/2004 07:14

181-200 chapter 8 BAT 18/11/2004 07:14

Annex 1. ETP MODEL CHARACTERISTICS

In order to quantitatively assess the merits of CO2 capture and storage technologies in comparisonto other technology and policy options, the IEA has used an in-house optimization tool known asthe Energy Technology Perspectives (ETP) model. This enables the benefits of CCS and othertechnology options, such as nuclear and renewable energy, to be cross-compared in a structured,logical and transparent manner.

This annex provides an overview of the structure and scope of the ETP model and the assumptionsthat lie behind the analysis. The general structure of the model is outlined, followed by a discussionof key technology parameters. The annex will be of interest to those wishing to understand theway in which the quantitative analysis has been structured in order to reach the results providedin Chapters 4, 5, 6 and 7.

The Value Added of the ETP Model CCS Analysis

The ETP model belongs to the MARKAL family of bottom-up systems engineering-economic models(Fishbone and Abilock, 1981; Loulou et al., 2004). MARKAL has been developed during the past30 years by the Energy Technology Systems Analysis Programme (ETSAP), one of the IEA implementingagreements (ETSAP, 2003).

A model is a structured, logical and reproducible method to analyse a complex policy problem.While no-one can predict the future with certainty, the goal is to ‘model for insights’, not to ‘modelfor figures’. Any model of this kind is a highly stylised representation of the world energy supplyand demand, based on a dataset that approximates the real world. Each model has its own uniquecharacteristics that affect the results and conclusions.

The ETP model is a complex model. Using it for the purpose of CCS analysis requires a significanteffort. This raises the issue why such a complex model is needed for proper assessment. The reasonis that this model accounts for a large number of key characteristics of the energy system that areof importance for long-term CCS decision making. If these characteristics were not accounted forthe outcome of the analysis would be different, and in all likelihood the outcome would be lessrelevant for decision making. The model has the following features, some of which are elaboratedon in more detail in this annex:

● The model represents the world split into 15 regions. This detailed representation of the world energysystem accounts for the specific characteristics of the energy system on a regional level, such asavailability of primary energy resources, acceptance of nuclear, and regional capital availability.

● The model optimizes the energy system for the next 50 years. Such a long-term perspective isneeded in order to assess the role of CCS properly.

● The model provides insights concerning the impact of deploying CCS on global fossil fuel andelectricity markets, an issue that has received limited attention so far.

● The model accounts for competing emission reduction strategies in certain sectors. For examplerenewables and energy efficiency options are considered as well as CCS.

● The model accounts for interactions on the systems level. For example, use of more fossil fuelswith CCS may result in less use of renewable energy. If this is the case, the marginal benefits interms of emissions reduction of using CCS are small. Therefore the assessment of CCS CO2 benefits

ANNEX 1. ETP MODEL CHARACTERISTICS 201

201-218 Annexe 1 17/11/2004 16:42


with ETP can result in a fundamentally different answer than a back-of-envelope calculationwhere, for example, coal-fired power plants with and without CO2 capture are compared.

● The model accounts for the difference between CO2 capture and storage cost (model input) andCO2 emission abatement cost (model output). CCS is evaluated based on emission abatement cost,which is a better measure for the cost and benefits of an emission mitigation strategy.

● The model contains a database of current technologies and emerging technologies. Thereforethe assessment of CCS is not only based on the technological characteristics of the current energysystem, but also on the characteristics of the future energy system. This is of key importancebecause the characteristics of the future energy system will be very different from the currentenergy system if CO2 policies are introduced.

● The model representation is based on detailed technology data. These data have a solid basisin engineering studies and scientific literature. This solid basis enables the identification oftechnology RD&D opportunities. This is a value added compared to econometric top-down modelswith a very aggregate representation of particular technology, that does not allow validation ofparticular technology development prospects and that does not allow identification of RD&Dopportunities.

● The model accounts for capital stock turnover. This is of importance for proper assessment ofthe transition to fossil fuels with CCS.

● The model represents electricity supply and demand in detail, accounting for the difference ofbase load and peak load plants and intermittency of renewables. The future annual electricityload curve is calculated by region, based on the demand for useful energy. The load of individualpower plants can be varied over the year and during the lifespan of the plant, as is the case inpractice. CHP is represented in detail, with a seasonal heat load curve. This detailed representationof the electricity system is of key importance for the assessment of CCS in the electricity sector.

● The model accounts for future demand for synfuels such as hydrogen and DME for thetransportation market; based on a detailed representation of transportation demand, competingtransportation technologies and fuel supply options. This allows for proper assessment of CCSpotentials in fuels supply.

● The model accounts for carbon leakage through industry relocation and changes in global energymarkets, if regional CO2 policies differ.

The Model Representation of the Energy System

The ETP model is a micro-economic representation of part of the world economy, divided into 15 regions.1

Only the energy part of the economy is modelled (i.e., the energy system). The energy system isrepresented as a set of interlinked markets in economic equilibrium. The model covers the productionof primary energy carriers, their conversion into final energy carriers such as gasoline, electricity andheat, and the conversion of final energy carriers into useful energy or so-called energy services, suchas lighting and transportation. This so-called energy system (Figure A1.1) is modelled as a set ofinterdependent technical product flows and processes.2 Various technologies can be used to generatecertain product flows, e.g., a number of coal and gas-fired power plant types for electricity production.The model includes a technology database with around 1,500 supply and demand side technologies.

1. The 15 regions considered in this study are: Africa, Australia/New Zealand, Canada, China, Central and South America (CSA),Eastern Europe, the Former Soviet Union (FSU), India, Japan, Mexico, Middle East, Other Developing Asia (ODA), South Korea, the USand Western Europe.2. Models of this type, which start from descriptions of technical options, are often called ‘bottom-up models’ as opposed to ‘top-down’models that start from a description of the economy as a whole.

201-218 Annexe 1 17/11/2004 16:42

ETP is a linear programming model that minimizes an objective function. This objective functionrepresents total discounted energy systems cost over a number of periods that satisfies a certainenergy demand under certain constraints (e.g., the attainment of certain production levels, theavailability of certain technologies, etc.). ETP is a partial equilibrium model: the model solutionrepresents the equilibrium that would be achieved in an ideal market and (according to neoclassicalwelfare economics) would maximise welfare. The model version that is used for this analysis has afixed demand for energy services. Other versions exist where the useful energy demand respondsto price changes. However, for this analysis this additional complexity is not included.

The technology choice and process activity levels in the model determine the physical and monetaryflows within the energy system. A model solution consists of a set of process activities, flows and resultingemissions (the so-called primal solution in linear programming), and prices (the so-called dual solution).

The strength of these types of model is that they are very well suited to assessing long-term investmentdecisions for complex systems, where future technology characteristics are very different from currenttechnology. This is in contrast with so-called top-down models that have little technological detail.Moreover, the single objective function ensures that the resulting scenario is internally consistent,as decision making for all processes and all flows is based on the same criteria.

On the other hand, these types of models have no explicit representation of macro-economic relations.Therefore the impact of changes in the energy system on labour markets and financial markets is nottaken into account. The analysis in this book is limited to the variations of the energy system.

Black boxes known as ‘processes’ or ‘technologies’ are the building blocks of a MARKAL model. Theyare characterized by:

● their physical inputs and outputs of energy;

● their costs;


Figure A1.1

The ETP model reference energy system

Hydrogenproduction

Industry

Residential/commercial


Refineries

Transport

Heating

Cooling

Power

Moving

etc.

Gasoline

Natural gas

Electricity

Coke

Hydrogen

Heat

etc.

Renewables

Fossil fuels

Nuclear

Usefulenergy

Primary energy

Conversion sectors/processes

Finalenergy

Demandsectors/processes

Coke ovens

Heatproduction

201-218 Annexe 1 17/11/2004 16:42


● other characteristics such as environmental impacts (in this study GHG emissions), over a numberof time periods.

Implicitly these process descriptions yield a very detailed input-output structure linking hundreds ofinterdependent processes through flows of materials and energy. The model covers all major processesand energy chains ‘from well to wheel’ (Figure A1.1). Given adequate input of data for the individualtechnologies, the model structure is suitable for life cycle analysis of both energy and materials in adynamic perspective. Upstream and downstream effects are taken into account.

Process descriptions follow a standard format, consisting of two data sheets. One sheet describes thephysical inputs and outputs (of energy and materials), while the other characterizes the economicand remaining process data. The input data structure depends to some extent on the process that ischaracterized. Data for fuel mining and transportation, power plants, other transformation processesin the energy sector, materials manufacturing industry, and other end-use technologies are characterizedin different units (e.g., per kW for power plants and per tonne product for materials-producing industries).A schematic example of the model input structure for power plants is shown in Table A1.1.

The data input is divided into eleven time periods. These cover the period 2000-2050, meaning thateach period represents five years. One column is reserved for time-independent variables (TID). Thephysical data refer to all the physical inputs and outputs that are considered relevant in this study;inputs and outputs of energy products and materials as well as emissions of all relevant GHG emissions.GHG include CO2, N2O, CH4, with their usual weights, corresponding to their 100-year global warmingpotential. The physical process data do not represent the total mass and energy balance where inputequals output (because of flows that are not accounted for, such as low temperature waste heat).

In order to keep track of costs under changing economic environments, the data sheet distinguishesbetween three cost categories:

● investment costs (which are proportional to the installed capacity),

● fixed annual costs (proportional to the installed capacity), and

● variable costs (proportional to production volume).

Regional cost multipliers and region specific discount rates are applied in order to reflect the differenteconomic conditions (see Annex 2).

Flexibility in the input/output ratiosThis input structure enables the representation of changing technology parameters in time. For instance,increasing process efficiency can be modelled by decreasing inputs per unit of output (such as forfuel in Table A1.1). Decreasing costs or changing restrictions can be modelled in a similar way. Thisis illustrated by the investment costs in Table A1.1 which decrease over time. This is one way to accountfor so-called ‘learning-by-doing’, accounting for decreasing costs as the installed capacity increases.The more complex option is where learning is a function of cumulative investments as calculatedby the model (so-called endogenous technology learning). This approach has not been applied inthis study.

BoundsThe user of the model can impose restrictions on the deployment of certain technologies. Suchrestrictions (called ‘bounds’ or constraints) may reflect consumer or political preferences, intentionsor objectives expressed in policy papers, or long- and short-term physical constraints such asnatural resource availability.

201-218 Annexe 1 17/11/2004 16:42

In this study the following types of bounds play a role:

● bounds on maximum penetration of certain technologies, reflecting social and strategicconsiderations for instance (e.g., a maximum bound on nuclear and hydropower, a maximumimport of natural gas from Russia). These bounds are mainly based on acceptance issues or ontechnical data concerning the availability of resources and the timespan necessary to startimplementing the required technologies (e.g., the necessary time for building pilot plants andplant construction);

● reflecting the standing capacity from earlier periods (e.g., for the existing building stock);

● bounds on the availability of natural resources (e.g., availability of oil, gas and renewable energy).

The ETP model matrix contains 700,000 rows and 750,000 columns, and 5 million non-zerocoefficients. Given the size of the model, it is not possible to discuss all the input data in detail.Only the general structure of the model will be discussed, followed later on in this chapter by adiscussion of key modules that affect the CCS technology choice.

Demand categories

Processes represent all activities that are necessary to provide certain products and services – for examplespace heating or vehicle-miles to be travelled. Many products and services can be generated througha number of alternative (sets of) processes that feature different costs and different GHG emissions.

The current model contains 106 ‘demand categories’ across the main end-use sectors. A generaloverview is provided in Table A1.2. For each demand category, energy service demands are specifiedin terms of useful energy or so-called energy services (e.g., vehicle kilometers).


Table A1.1

MARKAL model data structure for a power plant - an example

Period Unit TID 2000 2005 2010 … 2050

Sheet 1: Physical flowsInputs Fuel (GJ/GJel) 2.0 1.8 1.6 … 1.4

Output Electricity (GJel) 1 1 1 … 1

Sheet 2: Other dataInvestments (USD/kW) 1000 800 700 ... 600

Fixed annual costs (USD/kW.yr) 5 5 5 … 5

Variable costs (USD/GJel) 2 2 2 … 2

Delivery costs (USD/GJ) 1 1 1 … 1

Availability factor (unit/unit cap) 0.9 0.9 0.9 … 0.9

Peak contribution (kW/kW) 1 1 1 … 1

Life (years) 25

Start (year) 2000

CO2 emitted (kg/GJel) 15 14 13 … 10

CO2 captured (kg/GJel) 150 135 120 … 105

Residual capacity (GW) 2 0 0 … 0

Maximum capacity (GW) 5 10 50 … 50

Minimum capacity (GW) 0 0 0 … 0

201-218 Annexe 1 17/11/2004 16:42


As ETP is a global model, it covers trade in energy commodities and industrial commodities. Tradeis limited by cost only. In this approach it would be possible to account for carbon leakage effectsdue to changes in global commodity trade.

The BASE scenario GDP growth (see Annex 3) and electricity demand are calibrated with the2004 World Energy Outlook (IEA, 2004a), but it is virtually impossible to achieve a 100% match.ETP demand is defined in useful energy terms, and ETP final and primary energy demand is aresult of technological development, efficiency trends and cost optimization. On the other handWEO is an econometric model, where projected final energy demand is based on econometricdata. The very different nature of both modelling approaches will result in different outcomes.

The model includes a detailed database of energy supply and energy demand technologies. Onthe demand side, this database contains energy efficiency options and energy substitution options.For example a hybrid car is modelled as an alternative for a conventional gasoline-fuelled internalcombustion engine, while at the same time a hydrogen-fuelled fuel cell car is considered. The

Table A1.2

Demand categories in the ETP model

Sector Number of demand categories

Agriculture 1

Services 17

Power plants own use 1

Industry 46

Non-energy use 7

Residential 19

Transportation 15

Total 106

Future technology characteristics: a key uncertainty

An analysis with a broad time horizon (2050 in this study) will be based on technology datafrom different sources. These data will have different levels of accuracy. Often the accuracy ofthe data is not clear. Generally speaking, assessment studies for new technologies will oftensuggest a significant improvement potential compared to existing technologies. However, thedata are more uncertain. In fact, many new technologies do not make it to the market. In aleast-cost planning model with perfect foresight, such as ETP, uncertainty is not accounted for.

Without proper guidance by the modeler, risky speculative technologies are selected insteadof less attractive but proven technologies. Such technology optimism can create modellingresults that suggest radical technological change. Considering only proven technologies canincrease the credibility of the study. However, consideration of technological change maylead to radically different policy conclusions. Therefore, the model should contain a balanceddataset, and common sense is required regarding the conclusions that are drawn from anymodel run including speculative technologies. The technology dataset should be part ofthe uncertainty analysis.

201-218 Annexe 1 17/11/2004 16:42

technology choice depends on least-cost criteria that include regional fuel prices, discount ratesand technology cost assumptions.

The ETP model structure and model data have been characterized in more detail in a number ofpublications, e.g., (Gielen and Karbuz, 2003; Gielen et al., 2004).

The Fossil Fuel Supply Module

In most sectors the fuel prices constitute a set of key parameters that determine the fuel choice.The price assumptions from the World Energy Outlook are listed in Table A1.3. These prices havebeen used to calibrate the model. The figures indicate a coal and gas price gap in 2030 rangingfrom 0.60 USD/GJ in regions with ample gas resources up to 3.27 USD/GJ in regions with LNGimports and indigenous coal reserves.

Fuel prices in the ETP model are endogenous. This is a major difference with other bottom-up models,where fuel supply curves are defined exogenously. Using endogenized fuel prices enables the impactof CCS technology on fuel supply to be taken into account. Coal, gas and oil markets have beenmodelled. Understanding the model structure can help to understand the interaction of CCStechnology and fossil fuel markets. The fossil fuel supply model structure will therefore be discussedin more detail.

Figure A1.2 shows the oil production and processing module. Only one type of crude oil has beenmodeled. The numbers in the figure refer to the number of technologies in a specific category.Primary, secondary and tertiary oil production are modelled as a sequence of processes. Crude oilcompetes with syncrude and oil products compete with synthetic fuels. CO2 EOR competes withother methods for enhanced oil recovery.

There is consensus that oil reserves will not be exhausted over the next 50 years (IEA, 2001). Infact, total conventional oil production increases in the IEA WEO projections from 74 mbpd in2002 to 108 mbpd in 2030, which represents an increase of 49% (IEA, 2004a). Non-conventionaloil increases to 10.1 mbpd by 2030. This non-conventional oil production could increase further.The growth in oil demand will be largely met by producers in the Middle East.


Source: IEA, 2004a.

Table A1.3

Coal and gas price projections, 2000-2030

2000 2010 2020 2030

Oil (USD/GJ) 4.95 3.93 4.52 5.12

Gas USA/CAN/MEX/CSA (USD/GJ) 3.67 3.58 3.99 4.42

WEUR/EEUR/AUS (USD/GJ) 2.88 3.11 3.59 4.06

FSU/MEAST/AFR/OASIA (USD/GJ) 1.34 1.15 1.63 2.10

JAP/SKO/CHI/IND (USD/GJ) 4.48 3.66 4.13 4.52

Coal AUS/CHI/USA (USD/GJ) 1.00 1.05 1.10 1.15

Others (USD/GJ) 1.14 1.36 1.44 1.50

201-218 Annexe 1 17/11/2004 16:42


Fig

ure

A1

.2

Oil

and

gas

pro

duct

ion

mod

ule

Prob

able

/pos

sible

re

serv

esPr

oved

rese

rves

Incr

ease

d re

cove

ry

Prim

ary

pro

du

ctio

n o

nsh

ore

/off

sho

re (

10

)

Seco

nd

ary

pro

du

ctio

n o

nsh

ore

/off

sho

re (

10

)

Tert

iary

pro

du

ctio

n o

nsh

ore

/off

sho

re C

O2 b

ased

(7

)

Tert

iary

pro

du

ctio

n o

nsh

ore

/off

sho

re s

team

bas

ed (

7)

Tert

iary

pro

du

ctio

n o

nsh

ore

/off

sho

re p

oly

mer

/mic

ella

r (7

)

Gas

pro

du

ctio

nN

atu

ral G

as L

iqu

ids

Refin

ery

Pre-

trea

tmen

t

Pre-

trea

tmen

t

Pre-

trea

tmen

t

Oil

san

ds

op

en-c

ast/

un

der

gro

un

d (

2)

Hea

vy c

rud

e/Ta

r sa

nd

s

Oil

shal

e

Gas

/Co

al/B

iom

ass

to L

iqu

ids

FT

Gas

/Co

al/B

iom

ass

to M

eOH

/DM

E

Bio

mas

s to

EO

H

Bio

mas

s h

ydro

ther

mal

liq

uef

actio

n -

bio

die

sel

Not

e: F

igur

es in

bra

cket

s re

fer t

o th

e nu

mbe

r of t

echn

olog

ies

in a

spe

cific

cat

egor

y. FT

= F

isch

er-Tr

opsc

h; M

eOH

= M

etha

nol;

EOH

= E

than

ol.

201-218 Annexe 1 17/11/2004 16:42

In the model, increasing dependence on producers in the Middle East results in price increases.This is reflected in the model through a supply curve for producers in the Middle East. The highertheir production, the higher the price. This curve is split into 8 discrete steps. The ETP model supplycurve is based on the 2002 WEO (IEA, 2002a).

Figure A1.3 shows the ETP model structure for gas supply. A number of gas supply options havebeen considered. Associated gas has been considered as a single category together with conventionalgas. A number of unconventional supply options have been considered. For example in the USA,unconventional gas production already constitutes a significant share of total gas production.

Gas transportation constitutes a key cost component, so transportation pipelines and LNGtransportation have been modelled in detail. Stranded gas (at remote locations) and gas close toconsumer markets have been modelled separately. Stranded gas can be converted into synfuels orit can be converted into LNG. In the longer term new types of high-pressure pipelines may allowtransportation of gas from remote sites to consumer markets. Pipelines from the Middle East toEurope deserve special attention in this respect. For the time being, such pipelines have not beenconsidered. While pipeline supply is a suitable option for Europe and possibly for East Asia, the USwill increasingly rely on LNG imports. This results in higher regional gas prices.

The coal market is a competitive market, with many suppliers from around the world. Moreover, coalreserves are much more extensive than oil and gas reserves, so there is no strategic need for governmentsto intervene. Resource availability poses no constraints well beyond the model time horizon.


Figure A1.3

Structure for gas supply in the ETP model

Stranded gas

Conventional gas

Tight formation gas

Coalbed Methane

Geopressurized gas

Gas hydrates

Import pipeline

LNG import

LNG liquefactionshipping

Intra-regional pipeline

HP piplns from 2015

Export

Distribution ELE sector

Distribution IND sector

Distribution R&C sector

Distribution TRANS sector

Export Pipeline

Export

FT synthesis

Methanolproduction

201-218 Annexe 1 17/11/2004 16:42


Fig

ure

A1

.4

Stru

cture

for

coal

in

th

e ETP

mod

el

Bro

wn

coal

Was

hin

g

Sizi

ng

Rom

bro

wn

coal

Rom

ha

coal

Fin

esB

riq

uet

ting

Tran

spo

rtat

ion

(2

)

Tran

spo

rtat

ion

(2

)

Tran

spo

rtat

ion

(2

)

Pow

er p

lan

ts

Ind

ust

rial

bo

ilers

Pow

er p

lan

ts

Trea

ted

coal

Rom

la

coal

Tran

spo

rtat

ion

(2

)Re

sid

entia

l/C

om

mer

cial

Ind

ust

rial

bo

ilers

Tran

spo

rtat

ion

(2

)

Bit

um

inous

coal

,

hig

h a

sh

Bit

um

inous

coal

,

low

ash

Not

es: R

om =

Run

of m

ine;

ha

= hi

gh a

sh; l

a =

low

ash

201-218 Annexe 1 17/11/2004 16:42


The coal model structure is shown in Figure A1.4. Brown coal and hard coal have been modelledseparately. With regard to hard coal, raw coal and washed coal have been modelled separately forregions where the use of high-ash raw coal is significant. For all other regions only one coal type isdefined. It is assumed that all this coal is either washed or it is low-ash coal that needs no washing.The current model version does not account for the varying sulphur content and mercury content ofcoal. For the residential and commercial sector, briquettes have been modelled separately. Transportationhas been split into two categories (demand close to the mines and long-distance transportation).

The CCS Module

The ETP model structure for CCS in the electricity sector is shown in Figure A1.5. The structure canbe split into three parts: capture, transportation and storage. The set of capture technologies issplit into likely technologies (that are proven, or that require limited technology development) andspeculative technologies (whose demonstration on a relevant scale has not yet started). The likelytechnologies include supercritical power plants with flue gas capture, IGCC with fuel gas capture(for coal and for lignite), and gas-fired power plants with either flue gas or fuel gas CO2 capture.

The speculative technologies include chemical looping reactors for coal and gas, power plantsincluding solid oxide fuel cells for both coal and gas, and ultra-supercritical steam cycles for coal.The various capture technologies were discussed in detail in Chapter 3. For gas, the quality of thedataset does not allow a split of chemical absorption systems, pre-combustion natural gas reformingand oxyfueling; therefore all three have been represented by a single placeholder. Also, a numberof cogeneration units with CCS have been considered (cogeneration of heat and electricity andcogeneration of synfuels and electricity).

Several industrial, large-scale CHP technologies with CO2 capture have also been considered in themodel. These include biomass IGCC, black liquor IGCC, and gas-fired combined cycles for the chemicalindustry. The cogeneration units in other industries are smaller, making CO2 capture a less likelyoption. For iron and steel, CO2 capture from blast furnace gas has been considered; considerationof CCS for blast furnace gas-fired power plants would result in double counting.

Apart from CO2 capture in the electricity sector, capture in the manufacturing industry and in fuelsupply has been considered for the following processes:

In manufacturing:

● Blast furnaces; ● DRI plants;● Portland cement kilns;● Ammonia plants;

In fuels supply:

● Flexicoker;● Fischer-Tropsch synfuel production from gas, coal and biomass;● Hydrogen from gas, coal and biomass. Hydrogen use has been considered for all types of industrial

burners, steam boilers and kilns, and for refinery purposes. Also, hydrogen is considered as a fuel forresidential and commercial heating. Finally, hydrogen has been considered as a transportation fuel.

CO2 transportation costs have been varied for aquifers (3 USD/t), depleted oil and gas fields(10 USD/t), enhanced fossil fuel recovery (5 USD/t), and inter-regional CO2 shipping (15 USD/t).

201-218 Annexe 1 17/11/2004 16:42


These should be considered optimistic estimates for large-scale transportation systems that excludecost for pressurization. Variations in transportation costs within each category have not beenaccounted for. A number of CO2 storage options have been considered. Onshore and offshorepotentials have been characterized separately, as the acceptance for each may differ. The potentialsdiffer by region; some storage options are not available in certain regions.

The Renewables Module

For the accuracy of modelling, it is important to consider the regional techno-economic characteristicsand potentials of the renewable energy supply. A detailed Geographical Information System (GIS)has been developed for this purpose. The GIS data are aggregated into supply curves that serveas input for the global ETP model analysis. A GIS-analysis is based on the principle of overlayingmaps. It implies the use of location-specific data of high resolution for inherently consistent globalanalysis. This is important because, for many developing countries, accurate potential estimates arenot available.

Figure A1.5

Schematic CCS model structure for the electricity sector

Likely technologies

NGCCCA + membranes/fuel gas PA/oxyfueling

IGCC,CO2 removal fuel gas PA

SCSC,Chemical absorption/membranes

Onshore EOR

Depleted oil fields onshore

Depleted oil fields offshore

Onshore EGR

Offshore EGR

ECBM < 1 km depth

ECBM > 1 km depth

Onshore aquifers

Offshore aquifers

Methanol/DME production

Speculative technologies

IGCC+SOFC, coal fired

NGCC + chemical looping, gas fired

SCSC + chemical looping, coal fired

SOFC+CC, gas fired

USCSC, chemical absorption/membranes

Pipeline transportation/

shipping

Note: Only centralized fossil-fuelled power plants are shown in this figure.

SCSC = SuperCritical Steam Cycle. USCSC = Ultra Supercritical Steam Cycle. NGCC = Natural Gas fired Combined Cycle. CA = ChemicalAbsorption. PA = Physical Absorption. SOFC = Solid Oxide Fuel Cell. IGCC = Integrated Gasification Combined Cycle.

201-218 Annexe 1 17/11/2004 16:42

The GIS analysis produces estimates of regional supply curves of sufficient quality for the model,based on data sets publicly available, without the need of surveys of each individual region. It buildson meteorological and geological datasets. The assessment has focused on solar, wind andgeothermal. The meteorological data sets (ECMWF, 2003, and Czisch, 2003) are based on a 1°x1°global grid (i.e., 111x111 km at the equator).

Auxiliary data sets on land-cover, population distribution, topography, etc., have significantly higherresolution, usually 1, 5 or 15 minutes (1 minute = 1.8 km at the equator). An overview of key assumptionsis shown in Table A1.4. The resulting global capacity potentials should be compared to a projectedneed for 7000 GW electric capacity by 2030 (IEA, 2002). The renewables potentials do not pose aconstraint for their expansion in the electricity sector. Cost and intermittency limit the growth.


Table A1.4

Key assumptions in the GIS analysis of the potential for renewables

Wind onshore ● Certain land cover types have been excluded (no urban areas, forests or wetlands);

● The distance to the centres of demand has been considered; either within 25 km (cost class A)or within 100 km (cost class B);

● Dense population areas have been excluded because of noise and acceptance problems (morethan 100 persons/km2);

● A maximum of 4% of remaining area is considered exploitable in order to account for competingland uses and other acceptance and environmental concerns;

● The resulting area translated into potential using an average turbine density of 13 MW/km2;

● Resulting global potential capacity 8,252 GW;

● This is divided into a number of wind speed classes.

Wind offshore ● Electricity grid access within 50 km (cost class A) or within 100 km (cost class B);

● Water depth less than 25 m (cost class A) or less than 50 m (cost class B);

● 33% of remaining area is considered exploitable;

● The resulting area is translated into a wind potential using an average turbine density of 13 MW/km2;

● Resulting global potential capacity 5,597 GW;

● This is divided into a number of wind speed classes.

PV ● Access to electricity grid;

● 1% of remaining area exploitable (this accounts also for land cover limitations);

● Resulting area translated into potential using 40 Wp/m2 ;

● Resulting global potential capacity 19,482 GW.

Solar thermal ● Land cover (no urban areas, forests or wetlands);

● Access to electricity grid;

● Not too densely populated (less than 100 persons/km2);

● 0.5% of remaining area exploitable;

● Resulting area translated into potential using 40 Wp/m2;


Geothermal ● Cost split into drilling and above ground installations;

● Three types of reservoirs: high and low quality hydrothermal and hot dry rock;

● Total geothermal heat potential of 43 EJ/yr is distributed to the 15 ETP regions and subdividedinto heat flow classes using GIS heat flow data;

● Five heatflow classes have been defined with varying drilling cost;


201-218 Annexe 1 17/11/2004 16:42


The supply curve for hydropower is split into large dams, run of river and small hydropower. Largedams are split into six classes. Three cost classes are competitive at current cost levels, and threecost classes are technological potentials that are not yet cost competitive. Repowering of existinghydropower installations has been considered as a separate option that can increase capacity by15%. The potentials are based on a World Energy Council study (WEC, 2001). Data for smallhydro are taken from the recent IEA renewable electricity book (IEA, 2003b). The total hydro potentialis almost 60 EJ electricity per year (about 4500 GW).

Geothermal heat supply is split into three types: high and low quality hydrothermal reservoirs andhot dry rock. A GIS data set for geothermal heat flow (Pollack et al., 1991) is used to allocate theglobal potential to the 15 ETP regions, and to subdivide the potential into five heat flow classes.Each heat flow class results in a different heat temperature and has therefore a different electricityproduction efficiency, and different drilling cost. The total worldwide potential amounts to 43 EJgeothermal energy (Bertani, 2003). In electricity terms, the potential amounts to 13.3 EJ. Surfacecosts are set at 1,000 USD/kW, and drilling costs for the least-cost class amount to 460 USD/kWand increase up to 1,700 USD/kW for the most expensive cost class (Stefansson, 1999).

Biomass supply has been split into a range of by-products and waste biomass as well as dedicatedplantations. For example building waste, waste paper, forestry residues, production residues,commercial and non-commercial fuelwood have been considered. The total biomass potentialincreases from 70 EJ in 2020 to 150 EJ in 2050. This potential is uncertain. Estimates for 2050 rangefrom 20 to 450 EJ, depending on future agricultural productivity and food consumption trends(Hoogwijk and Berndes, 2000).

In the ETP model, biomass co-combustion can take place in ordinary coal-fired and, following gasificationand gas cleaning, in gas-fired power plants. Dedicated biomass gasifiers have also been consideredin the electricity sector. In other sectors, there are options such as biomass combustion and black liquorgasification in the pulp and paper industry, and biocrude and bioalcohol production processes.

Figure A1.6 shows as an example the supply curve for onshore and offshore wind in Western Europe.Costs decline by about 25% in a period of 30 years due to technology learning. The potential forwind energy is substantial in terms of gigawatts, but the capacity factor for these plants iscomparatively low, between 20-40%, while fossil-fuelled plants can achieve a capacity factor of upto 95%. Even so, with about 800 GW total electricity capacity potential, wind can play a veryimportant role. The current incentives for wind energy in various European countries are also indicatedin this figure. Even higher incentives occur. Clearly these incentives are sufficient to achieve asubstantial growth in wind use. But the introduction of significant amounts of intermittent electricitysources such as wind raises questions about the security of electricity supply.

The intermittency of renewable electricity supply may cause problems because backup capacity isneeded in order to meet the demand during periods when insufficient renewable electricity is available.In practice, this usually implies the installation of gas turbines or oil fuelled engines. An alternativeapproach would be an electricity supply from renewables that is tailored for peak demand withsignificant amounts of surplus electricity used for production of hydrogen in periods of excess supply.The latter approach is considered in the ETP model, but it is rather energy inefficient and costly.

In regions where natural gas is not available, more capital intensive coal-fired units, pumpedhydro-storage units, or even nuclear may be operated as backup units. The investment cost of low-cost peaking units is in the range of 200-500 USD/kW but fossil fuel supply systems are neededfor these installations. These costs must be added to the cost of renewable systems. Another wayof dealing with peak load is demand management.

201-218 Annexe 1 17/11/2004 16:42

In a MARKAL model, intermittency is taken into account through a so-called peaking equation.The year is split into three seasons (winter, summer and intermediate) and day and night (a totalof six time slices). In the multi-regional ETP model, the split of seasons and day/night is the sameacross all regions.3 Winter and summer each represent three months while spring and autumn arerepresented by a single intermediate season. In the intermediate season, day and night represent12 hours each. In winter, day lasts 9.6 hours and night lasts 14.4 hours. In summer these are reversed.

For each demand category, the demand can be distributed over the six time slices. The shape ofthe electricity demand curve is calculated by the model as the sum of all demand categories (seeFigure A1.7). As the demand structure differs by region, the shape of the load curve differs by region.Reserve capacity accounts for fluctuating demand within one season. This reserve capacity is definedas a share of the demand in the time slice with maximum electricity load. It includes the capacityrequired to meet peak requirements in that time slice, forced outage, and scheduled outage. In theETP model, this capacity is set at 30%. The peaking equation ensures that the installed electricitycapacity equals demand in the time slice of maximum demand, plus the reserve capacity.

A peaking contribution is defined for each electricity supply technology. The peaking contributionranges from zero to 100%, and can be time slice specific (e.g., solar is only available during the day).For wind and for solar the peaking contribution has been set at 30% of the installed capacity, whileit is 100% for fossil-fuelled plants. In the model this peaking contribution has a certain monetaryvalue that is region and period specific. Usually it amounts to 10-15% of the electricity price.


Figure A1.6

ETP wind electricity supply curve for Europe

Key point: There is no single ‘true’ figure for renewables.The marginal supply cost depends on the quantity involved and on learning effects

USD

/kW

h

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0 200 400 600 800 1000

2000

2030

Price range current electricity production

Renewables feed-in tariff

GW

3. This is obviously a simplification. In principle the MARKAL code could be extended to allow for more detailed modelling of theelectricity load curve.

201-218 Annexe 1 17/11/2004 16:42


A second issue is the interaction between technology learning and investment costs. In practice,investments can result in technology learning, which results in cost reductions. The investment costsdecline by a fixed fraction for each doubling of the installed cumulative capacity (IEA, 2000). A‘virtuous cycle’ can occur, where additional investments result in additional learning which resultsin additional investments etc. This phenomenon can explain the rapid switch from one systemsconfiguration to another in reality or in models with endogenous technology learning.

Learning can be split into learning by R&D (innovation) and learning-by-doing. So far, it is notclear how cost-effective each can be, or which approach should be followed. In this analysis, a simpleapproach is applied that is in line with other bottom-up energy modelling studies. Emergingtechnologies are modelled explicitly as discrete technologies. This is a way of representing learningby innovation.

Learning-by-doing is important for technologies that start from a very low cumulative capacity, yetthe potential for fundamentally different process designs is limited. Learning-by-doing has been

Figure A1.7

ETP electricity load curves for Western Europe

Key point: Electricity demand varies over the year. Such variability may increase in the future.

Ele

pro

du

ctio

n in

dex

(-)

20002010

20202030

20402050

ID

SDIN

WDSN

WN

0

20

40

60

80

100

120

140

Note: The gap between peak and base load is a factor of two and it increases with time. Normalized to the daytime electricity demandin the intermediate season. Excludes reserve capacity. ID = Intermediate Day; IN = Intermediate dayS = Summer; W = Winter.

201-218 Annexe 1 17/11/2004 16:42

considered for all key renewables technologies4 (see chapter 5). The learning rate assumptions arecrucial to the cost reduction potential. It is difficult to estimate learning rates. True production costdata are scarce; often only price data are available that obscure cost reductions. Spill-over effectsfrom investments in other countries can affect the learning rates that are measured in a specificcountry.

Factors that commonly complicate the accurate projection of learning rates are: new technologieswith little or no price/cost history (e.g., PV, fuel cells); technologies with highly site-specific installationcosts (e.g., hydropower, biomass, geothermal); and technologies where market dynamics obscurethe relationship between capacity and investment costs (e.g., PV, combined cycle gas turbines).

In general, lower learning rates are found for technologies based on established, inherently large-scale components such as steam turbines. This is the case for fossil-fuelled power plants with CCStechnology. Higher learning rates occur for new designs with modular components suitable for mass-production manufacturing, such as PV modules, wind turbines and fuel cells (Neij, 1997).

Only CO2 benefits are considered in this analysis. Other pollutants represent about a quarter oftotal damage cost of electricity produced from fossil fuels. They are not taken into account inthis analysis (Table A1.5). This results in a slight underestimation of the benefits of renewable andnuclear electricity supply options, in comparison to CCS. The impacts and the valuation of theseimpacts are location specific and the emissions depend on the technology choice. Generally,abatement costs are much lower than damage costs. Another reason to exclude local air pollutantsis that many renewable energy options have other impacts that are hard to measure, such as horizonpollution by wind turbines or biodiversity impacts of biomass plantations.

The Nuclear Energy Module

Approximately 438 commercial nuclear generating units were in operation throughout the worldin 2000, with a total production capacity of 351 GW. Nuclear power is a CO2-free energy source.However, its use is controversial. The accidents at Three Mile Island and Chernobyl have virtuallystopped new investment in most OECD countries.

The problem of long-term storage of nuclear waste and the potential use of nuclear waste forproduction of dirty bombs or even nuclear bombs have prevented rapid growth in developing


4. One could argue that for example in the case of solar electricity, fundamental technological change can be of similar importance. Forexample, thin film technology and polymer PV systems are fundamentally different from the established crystalline and amorphous PVcells. Given the scope of this analysis, such issues were not taken into account.

Note: Figures refer to typical European power plants. CO2 emissions are valued at 50 USD/t CO2.

Source: Zwaan and Rabl, 2004.

Table A1.5

Damage costs for fossil-fuelled power plants

(US cents/kWh) PM10 SO2 NOx CO2 Total

Coal (built since 2000) 0.1 0.4 1.0 3.9 5.4

Oil (built since 2000) 0.1 0.5 1.2 2.9 4.8

Gas (built since 2000) 0.0 0.0 0.4 1.8 2.2

201-218 Annexe 1 17/11/2004 16:42


countries. On the other hand, growth in Asia continues, as countries such as Japan, China andIndia build new nuclear plants. Last year Finland announced the construction of a new nuclearreactor. Other European countries may follow. The development of new inherently safe reactor typessuch as the pebble bed modular reactor may reduce investment cost while eliminating both therisk of nuclear accidents and the problem that current reactors are only economic at a 1-2 GWscale (Kenny, 2004).

Nuclear energy faces a number of challenges (Rothwell and Zwaan, 2003). First, is the highcost of nuclear electricity, compared to fossil-fuelled power plants. Second, is the restrictedproliferation of the technologies on which it relies. Third, is the (perceived) risk of seriousaccidents, and fourth is the waste issue.

The cost structure of nuclear power plants is different from those of fossil-fuelled power plants ina number of ways. Investment costs dominate the cost profile, while operating costs are comparativelysmall. For an amortized French nuclear reactor, the production cost are 14 Mils/kWh (Bataille andBirraux, 2003). A recent MIT study states that based on numbers from ‘actual experience’ insteadof engineering projections, new nuclear electricity costs 6.7 US cents/kWh, at a real discount rateof 8.5%. Plausible but unproved reductions in capital and operating costs could lower that to 5.1 UScents/kWh (MIT, 2003). This is still 10-20% higher than for coal and gas-fired power plants withCO2 capture (these electricity production cost are discussed in Chapter 3).

The time needed to build and commission nuclear power plants is substantial. The interest on theworking capital during this period adds to the investment cost. Also, the costs of decommissioning,reprocessing and waste management are not negligible. In France, decommissioning costs are setat 15% of the cost of a nuclear reactor. Reprocessing of spent fuel and waste management representa similar provision (Economist, 2004). Therefore 30% of the costs are in fact not direct investmentcosts. If there are significant delays in the plant construction, the gap between overnight investmentcost and actual life-cycle investment cost will increase further. The costs of reprocessing are substantial;a once-through system without reprocessing results in a lower capital cost but higher waste volumes.

A number of new reactor designs are being studied that may reduce capital cost. For a series often European Pressurized water Reactors (EPRs) of 900 MW each, the investment costs are estimatedat 2,000 USD/kW (Bataille and Birraux, 2003). It is not clear what is included in this cost estimate,but it is in line with the overnight construction cost given by MIT (2003). In this study, a 25%reduction in these costs is projected. Adding decommissioning and waste fuel processing costs yields2,000 USD/kW as an optimistic estimate of future nuclear power plant cost.

A number of other designs are being proposed by different suppliers around the world. The lowestcost claim is for the Pebble Bed Modular Reactor, being developed by Eskom, at 1,400 USD/kW.This type of reactor is not yet proven on a commercial scale and this estimate probably excludesthe cost adders. Cost reductions for nuclear reactors can be achieved via standardized designs, serialproduction, economies of scale, and reduced construction periods. Given the importance of capitalcost, the discount rate at which nuclear power is evaluated is a key parameter. In a liberalizedmarket, nuclear has a disadvantage.

In the reference model calculations, nuclear capacity has been fixed in line with WEO ReferenceScenario assumptions. In a sensitivity analysis, the potential for nuclear was assessed, assumingcompetition with other emission mitigation strategies on a cost basis. The costs of new nuclearreactors were set at 2,200 USD per kW, declining to 2,000 USD per kW in the long term. Thisincludes spent fuel processing and decommissioning costs.

201-218 Annexe 1 17/11/2004 16:42

ANNEX 2. REGIONAL INVESTMENT COSTS AND DISCOUNT RATES 219

Table A2.1

Region specific cost multipliers

INVCOST FIXOM VAROM

AFR 125 90 85AUS 125 90 90CAN 100 100 100CHI 90 80 80CSA 125 90 85EEU 100 90 85FSU 125 90 85IND 90 80 80JPN 140 100 100MEA 125 90 85MEX 100 90 90ODA 125 80 80SKO 100 90 90USA 100 100 100WEU 110 100 95

5. These multipliers do not apply to energy and materials inputs that are modelled as physical flows. The regional price of these flowsis calculated by the model.

USA = 100.

Annex 2. REGIONAL INVESTMENT COSTS AND DISCOUNT RATES

Regional Investment Costs

The ETP model covers 15 regions. The database is set up as one ‘reference database’ with cost data forthe USA. Costs in other regions are calculated by multiplying US cost data with a region-specific factor.Region specific cost multipliers are listed in Table A2.1.5 These multipliers are applied to all processes.

This detailed, but still rather crude, representation of the world energy system poses certain limitations:

● The currency exchange rates tend to fluctuate. Changing exchange rates affect the relative investmentcosts. In particular, exchange rates for developing countries can fluctuate by a factor of two.

● The project system boundaries may differ by region and by site. For example in developing countriesit may be necessary to build roads, new power lines or other infrastructure for new power plants.

● The regions in the model are very large. Any cost factor is an average that may differ considerablyfor locations (and countries) within regions.

● Particularly in developing countries, some technologies require imported equipment, while othersare based on locally produced equipment. Such a difference can impact investment cost significantly.

● In developing countries the availability of skilled labour may be a limiting factor. If workershave to be hired from abroad, this will affect labour cost. Operating and maintenance costsconsist of 50% labour costs (that are region specific) and 50% materials and auxiliaries costs(that are assumed to be the same in all regions).

219-220 Annexe 2 18/11/2004 11:44


Discount Rates: Liberalization, Risk and TimePreferences

The discount rates in the model differ by region and by sector. Model discount rates, shown inTable A2.2, should reflect the real world discount rates. These discount rates are usually significantlyhigher than the long-term social discount rate. Economists’ opinions differ as to which discountrates should be applied for CO2 policy analysis (Portney and Weyant, 1999).

ETP model discount rates are real discount rates, excluding inflation. The discount rate will differamong world regions, depending on capital availability and perceived risk.

Money supply can be divided into loans and own capital and equity. The long-term return oninvestment for equity is several percent higher than for loans, because the owner of the equity isexposed to an increased risk (that the company goes bankrupt, in which case loans are paid backfirst, and usually the equity owner gets nothing). In a situation where electricity supply is governedby government, the lending rate may apply.

In a liberalized market, the equity rate is more plausible. The ETP figures are based on the 30-yeargovernment bond rate (for the main country in the region, if applicable), corrected for inflation.For developing countries Moody’s country ranking has been used as a measure for creditworthiness.Industry financing has been split into lending and equity (stocks etc.). One percentage point hasbeen added in the case of borrowing by companies, compared to government bond rates, in orderto reflect the average incremental risk associated with lending to companies. 5.5% has beenadded to the bond rate for industrial equity risk (NYU Stern, 2002).

Table A2.2

Region and sector specific discount rates in the ETP model

Real bond yield Industry/Electricity Industry/Electricity

2000-2001 (%/yr) Lending (%/yr) Equity (%/yr)

AFR 8.2 9.2 13.7AUS 2.6 3.6 8.1CAN 3.7 4.7 9.3CHI 5.2 6.2 10.7CSA 7.2 8.2 12.7EEU 5.7 6.7 11.3FSU 8.7 9.7 14.3IND 8.0 9.0 13.5JPN 2.0 3.0 7.5MEA 5.6 6.6 11.1MEX 7.2 8.2 12.7ODA 8.2 9.2 13.7SKO 5.6 6.6 11.1USA 4.2 5.2 9.7WEU 3.7 4.7 9.3

219-220 Annexe 2 18/11/2004 11:44

Annex 3. GDP PROJECTIONS

Gross domestic product (GDP) growth is a key driver for future emissions and, therefore, for thepotential of CCS technologies. The GDP projections in the ETP model’s GLO50 reference scenarioare in line with the IEA World Energy Outlook 2004 (IEA, 2004a). The growth projections byperiod and by region are shown in Table A3.1.

Table A3.2 shows what these growth figures mean in per capita GDP on a purchasing power parity(PPP) basis. The figures suggest a strong convergence of income levels. In 2050 the poorest worldregions will reach the same per capita GDP as Europe had in 2000. Obviously this is an optimisticassumption that implies high economic growth in developing countries for the next half century.In sensitivity analysis, the impact of lower and higher growth rates was investigated. In particular,growth in developing countries has been varied (Tables A3.3 and A3.4).

ANNEX 3. GDP PROJECTIONS 221

Table A3.1

BASE/GLO50 GDP growth projections (% per year)

Averagegrowth

2000- 2005- 2010- 2015- 2020- 2025- 2030- 2035- 2040- 2045- 2000-2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 2050

AFR 3.6 3.6 3.6 3.6 3.6 3.6 3.6 3.6 3.6 3.6 3.6

AUS 1.1 1.5 1.4 1.4 1.4 1.4 1.5 1.5 1.5 1.5 1.4

CAN 2.3 2.7 2.3 2.1 2.0 1.9 1.7 1.7 1.7 1.8 2.0

CHI 7.1 5.8 5.2 4.6 4.1 3.8 3.5 3.0 2.5 2.0 4.1

CSA 2.0 3.5 3.3 3.1 2.9 2.7 2.5 2.5 2.5 2.5 2.7

EEU 3.4 3.4 3.4 3.4 3.5 3.5 2.4 2.4 2.4 2.4 3.0

FSU 3.1 3.1 3.1 3.1 3.1 3.1 2.5 2.5 2.5 2.5 2.9

IND 5.3 5.4 5.0 4.6 4.2 3.9 3.6 3.0 3.0 3.0 4.1

JPN 1.1 0.6 0.9 1.0 0.4 0.4 1.5 1.5 1.5 1.5 1.0

MEA 2.6 2.6 2.6 2.6 2.6 2.6 2.6 2.6 2.6 2.6 2.6

MEX 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4

ODA 4.0 4.1 4.0 3.9 3.9 3.9 3.2 3.1 2.6 2.6 3.5

SKO 4.0 4.4 3.9 3.3 2.9 2.6 1.3 0.9 0.8 0.4 2.4

USA 2.5 2.8 2.4 2.1 1.8 1.6 1.3 1.3 1.3 1.3 1.8

WEU 1.7 2.4 2.1 1.9 1.7 1.6 1.4 1.4 1.4 1.4 1.7

Global 3.2 3.5 3.2 3.1 2.9 2.7 2.5 2.4 2.3 2.1 2.8

221-224 Annexe 3 18/11/2004 11:45


Table A3.2

BASE/GLO50 per capita GDP

USD(2000)/capita 2000 2010 2020 2030 2040 2050

OECD North America 26.0 30.4 34.1 38.0 41.8 45.3OECD Europe 18.8 23.2 28.3 33.2 40.0 47.7OECD Pacific 22.1 26.5 32.4 39.3 47.3 55.9FSU 5.6 7.5 10.1 13.5 16.5 20.1Eastern Europe 4.6 6.4 9.1 12.8 16.4 21.0China 3.8 6.1 9.8 15.6 22.7 32.9Other Asia 3.3 4.7 6.7 9.6 13.6 19.4India 2.2 3.5 5.5 8.7 12.3 17.6Middle East 5.7 7.3 9.5 12.3 15.9 20.5Latin America 6.3 8.4 11.3 15.2 19.5 25.0Africa 1.9 2.7 3.9 5.6 7.9 11.3

Table A3.3

GDP growth projections for the sensitivity analysis with lower growth rates (% per year)

Averagegrowth

2000- 2005- 2010- 2015- 2020- 2025- 2030- 2035- 2040- 2045- 2000-2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 2050

AFR 3.6 3.6 3.1 3.1 3.1 2.6 2.6 2.6 2.6 2.6 3.0

AUS 1.1 1.5 1.2 1.2 1.2 0.9 1.0 1.0 1.0 1.0 1.1

CAN 2.3 2.7 2.1 1.9 1.7 1.4 1.2 1.2 1.2 1.3 1.7

CHI 7.1 5.8 4.2 3.6 3.1 2.3 2.0 1.5 1.0 0.5 3.1

CSA 2.0 3.5 2.8 2.6 2.4 1.7 1.5 1.5 1.5 1.5 2.1

EEU 3.4 3.4 2.9 2.9 3.0 2.5 1.4 1.4 1.4 1.4 2.4

FSU 3.1 3.1 2.6 2.6 2.6 2.1 1.5 1.5 1.5 1.5 2.2

IND 5.3 5.4 4.5 4.1 3.7 2.8 2.6 2.0 2.0 2.0 3.4

JPN 1.1 0.6 0.7 0.9 0.3 -0.1 1.0 1.0 1.0 1.0 0.7

MEA 2.6 2.6 2.1 2.1 2.1 1.6 1.6 1.6 1.6 1.6 1.9

MEX 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4

ODA 4.0 4.1 3.5 3.4 3.4 2.9 2.1 2.1 1.6 1.6 2.9

SKO 4.0 4.4 3.9 3.3 2.9 2.6 1.3 0.9 0.8 0.4 2.4

USA 2.5 2.8 2.2 1.9 1.6 1.1 0.8 0.8 0.8 0.8 1.5

WEU 1.7 2.4 1.9 1.7 1.5 1.1 0.9 0.9 0.9 0.9 1.4

Global 3.2 3.5 2.8 2.6 2.4 1.8 1.6 1.5 1.3 1.2 2.2

221-224 Annexe 3 18/11/2004 11:45

ANNEX 3. GDP PROJECTIONS 223

Table A3.4

GDP growth projections for the sensitivity analysis with higher growth rates(% per year)

Averagegrowth

2000- 2005- 2010- 2015- 2020- 2025- 2030- 2035- 2040- 2045- 2000-2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 2050

AFR 3.6 3.6 3.8 3.8 3.8 3.8 3.8 3.8 3.8 3.8 3.8

AUS 1.1 1.5 1.5 1.5 1.5 1.5 1.6 1.6 1.6 1.6 1.5

CAN 2.3 2.7 2.5 2.3 2.2 2.4 2.2 2.2 2.3 2.3 2.3

CHI 7.1 5.8 5.7 5.1 4.6 4.8 4.5 4.0 3.5 3.0 4.8

CSA 2.0 3.5 3.5 3.3 3.1 3.2 3.0 3.0 3.0 3.0 3.1

EEU 3.4 3.4 3.9 3.9 4.0 4.5 3.4 3.4 3.4 3.4 3.7

FSU 3.1 3.1 3.6 3.6 3.6 4.1 3.5 3.5 3.5 3.5 3.5

IND 5.3 5.4 5.5 5.1 4.7 4.9 4.6 4.0 4.0 4.0 4.8

JPN 1.1 0.6 1.0 1.1 0.5 0.5 1.6 1.6 1.6 1.6 1.1

MEA 2.6 2.6 3.1 3.1 3.1 3.6 3.6 3.6 3.6 3.6 3.3

MEX 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4 3.4

ODA 4.0 4.1 4.5 4.4 4.4 4.9 4.2 4.2 3.6 3.6 4.2

SKO 4.0 4.4 3.9 3.3 2.9 2.6 1.3 0.9 0.8 0.4 2.4

USA 2.5 2.8 2.6 2.3 2.0 2.1 1.8 1.8 1.8 1.8 2.1

WEU 1.7 2.4 2.2 2.0 1.8 1.8 1.6 1.6 1.6 1.6 1.8

Global 3.2 3.5 3.5 3.4 3.2 3.4 3.2 3.1 3.0 2.9 3.2

221-224 Annexe 3 18/11/2004 11:45

221-224 Annexe 3 18/11/2004 11:45

Annex 4. WEBSITES WITH MORE INFORMATION ON CCS

The following websites provide a starting point for those wanting more information on CO2 captureand storage technologies. The list is not exhaustive and some of the websites may have a limitedlifespan.

Second Annual Conference on Carbon Sequestration, Alexandria, VA,:www.carbonsq.com/proceedings.cfm

Third Annual Conference on Carbon Sequestration, Alexandria, VA:www.carbonsq.com/

CO2 capture project (activity of eight leading energy companies):www.co2captureproject.org/index.htm

IEA GHG R&D Programme: www.ieagreen.org.uk/

IEA GHG Programme R&D project database:http://script3.ftech.net/~ieagreen/co2sequestration.htm

NOVEM overview of CCS projects: www.cleanfuels.novem.nl/projects/international.asp

DOE carbon sequestration website: http://carbonsequestration.us/

Natural Resources Canada CO2 capture and storage roadmap:www.nrcan.gc.ca/es/etb/cetc/combustion/co2trm/htmldocs/technical_reports_e.html

Innovative technlogieen zur Stromerzeugung – auf dem weg zu CO2-freien Kohle – undGaskraftwerken. Conference proceedings, May 10-12 2004, Berlin.www.kraftwerkskongress.de/deu/index.htm

IEA Clean Coal Centre: www.iea-coal.co.uk/site/index.htm

Carbon sequestration leadership forum: www.cslforum.org/

ANNEX 4. WEBSITES WITH MORE INFORMATION ON CCS 225

225-226 Annexe 4 18/11/2004 11:47

225-226 Annexe 4 18/11/2004 11:47

Annex 5. DEFINITIONS, ABBREVIATIONS, ACRONYMSAND UNITS

This section provides definitions of the energy, economic and financial terms and the regionalgroupings used throughout this publication.

Fuel and Process Terms

Readers interested in obtaining more detailed information should consult the annual IEA publicationsEnergy Balances of OECD Countries, Energy Balances of Non-OECD Countries, Coal Information,Oil Information, Gas Information and Electricity Information.

API Gravity

Specific gravity measured in degrees on the American Petroleum Institute scale. The higher the number,the lower the density. 25 degrees API equals 0.904 kg/m3. 42 degrees API equals 0.815 kg/m3.

Aquifer

An underground water reservoir. If the water contains large quantities of minerals it is a saline aquifer.

Associated Gas

Natural gas found in a crude oil reservoir, either separate from or in solution with the oil.

Biomass

Biomass includes solid biomass and animal products, gas and liquids derived from biomass, industrialwaste and municipal waste.

Coal

Unless stated otherwise, coal includes all coal: both coal primary products (including hard coaland lignite) and derived fuels (including patent fuel, coke-oven coke, gas coke, coke-oven gas andblast-furnace gas). Peat is also included in this category.

Electricity Production

Electricity production shows the total amount of electricity generated by power plants. It includesown-use and transmission and distribution losses.

Enhanced Coal-bed Methane Recovery (ECBM)

ECBM is a technology for recovery of methane (natural gas) through CO2 injection into uneconomiccoal seams. The technology has been applied in a demonstration project in the US, and is beingtested elsewhere.

Enhanced Gas Recovery (EGR)

EGR is a speculative technology where CO2 is injected into a gas reservoir in order to increase thepressure in the reservoir, so more gas can be extracted.

ANNEX 5. DEFINITIONS, ABBREVIATIONS, ACRONYMS AND UNITS 227

227-234 Annexe 5 17/11/2004 16:42


Enhanced Oil Recovery (EOR)

EOR is also known as tertiary oil recovery. It follows primary recovery (oil produced by the naturalpressure in the reservoir) and secondary recovery (using water injection). Various EOR technologiesexist such as steam injection, hydrocarbon injection, underground combustion and CO2 flooding.

Fischer-Tropsch (FT) synthesis

Catalytic production process for synthetic oil products. Natural gas, coal and biomass feedstockscan be used.

Fuel cell

A device which can be used to convert hydrogen into electricity. Various types exist that can beoperated at temperatures ranging from 80°C to 1,000°C. Their efficiency ranges from 40-60%.For the time being their application is limited to niche markets and demonstration projects due tohigh cost and the immature status of the technology, but their use is growing fast.

Gas

Gas includes natural gas (both associated and non-associated with petroleum deposits but excludingnatural gas liquids) and gas works gas.

Heat

In the IEA energy statistics, heat refers to heat produced for sale. Most heat included in this categorycomes from the combustion of fuels, although some small amounts are produced from electrically-powered heat pumps and boilers.

Hydro

Hydro refers to the energy content of the electricity produced in hydropower plants assuming 100%efficiency.

Integrated Gasification Combined Cycle (IGCC)

IGCC is a technology where a solid or liquid fuel (coal, heavy oil or biomass) is gasified, followedby electricity generation in a combined cycle. It is widely considered a promising electricity generationtechnology due to its potential for high electric efficiency and low emissions.

Liquefied Natural Gas (LNG)

LNG is natural gas which has been liquefied by reducing its temperature to minus 162 degreesCelsius at atmospheric pressure. In this way, the space requirements for storage and transport arereduced by a factor over 600.

Non-conventional Oil

Non-conventional oil includes oil shale, oil sands-based extra-heavy oil and bitumen and derivativessuch as synthetic crude products, and liquids derived from natural gas (GTL).

Nuclear

Nuclear refers to the primary heat equivalent of the electricity produced by a nuclear plant withan average thermal efficiency of 33%.

227-234 Annexe 5 17/11/2004 16:42

Oil

Oil includes crude oil, natural gas liquids, refinery feedstocks and additives, other hydrocarbons andpetroleum products (refinery gas, ethane, liquefied petroleum gas, aviation gasoline, motor gasoline,jet fuel, kerosene, gas/diesel oil, heavy fuel oil, naphtha, white spirit, lubricants, paraffin waxes,petroleum coke and other petroleum products).

Other Renewables

Other renewables include geothermal, solar, wind, tide, and wave energy for electricity generation.Direct use of geothermal and solar heat is also included in this category.

Other Transformation, Own Use and Losses

Other transformation, own use and losses covers the use of energy by transformation industriesand the energy losses in converting primary energy into a form that can be used in the finalconsuming sectors. It includes energy use and loss by gas works, petroleum refineries, coal and gastransformation and liquefaction. It also includes energy used in coal mines, in oil and gas extractionand in electricity and heat production. Transfers and statistical differences are also included in thiscategory.

Renewables

Renewables refer to energy resources, where energy is derived from natural processes that arereplenished constantly. They include geothermal, solar, wind, tide, wave, hydropower, biomass, andbiofuels.

Purchasing Power Parity (PPP)

The rate of currency conversion that equalizes the purchasing power of different currencies, i.e.,makes allowance for the differences in price levels between different countries.

Scenario

An analysis dataset based on a consistent set of assumptions.

REGIONAL GROUPINGS

Africa

Comprises: Algeria, Angola, Benin, Botswana, Burkina Faso, Burundi, Cameroon, Cape Verde, theCentral African Republic, Chad, Congo, the Democratic Republic of Congo, Cote d’Ivoire, Djibouti,Egypt, Equatorial Guinea, Eritrea, Ethiopia, Gabon, Gambia, Ghana, Guinea, Guinea-Bissau, Kenya,Lesotho, Liberia, Libya, Madagascar, Malawi, Mali, Mauritania, Mauritius, Morocco, Mozambique,Niger, Nigeria, Rwanda, Sao Tome and Principe, Senegal, Seychelles, Sierra Leone, Somalia, SouthAfrica, Sudan, Swaziland, the United Republic of Tanzania, Togo, Tunisia, Uganda, Zambia andZimbabwe.

Central and South America

Comprises: Antigua and Barbuda, Argentina, Bahamas, Barbados, Belize, Bermuda, Bolivia, Brazil,Chile, Colombia, Costa Rica, Cuba, Dominica, the Dominican Republic, Ecuador, El Salvador,


227-234 Annexe 5 17/11/2004 16:42


French Guiana, Grenada, Guadeloupe, Guatemala, Guyana, Haiti, Honduras, Jamaica, Martinique,Netherlands Antilles, Nicaragua, Panama, Paraguay, Peru, St. Kitts-Nevis-Anguilla, Saint Lucia, St.Vincent-Grenadines and Suriname, Trinidad and Tobago, Uruguay and Venezuela.

China

Refers to the People’s Republic of China.

Developing Countries

Comprises: China, India and other developing Asia, Central and South America, Africa and theMiddle East.

Eastern Europe

Comprises: Albania, Bosnia-Herzegovina, Bulgaria, Croatia, Macedonia, Poland, Romania, Slovakia,Slovenia, Yugoslavia.

Former Soviet Union (FSU)

Comprises: Armenia, Azerbaijan, Belarus, Estonia, Georgia, Kazakhstan, Kyrgyzstan, Latvia, Lithuania,Moldova, Russia, Ukraine, Uzbekistan, Tajikistan, Turkmenistan.

Middle East

Comprises: Bahrain, Iran, Iraq, Israel, Jordan, Kuwait, Lebanon, Oman, Qatar, Saudi Arabia, Syria,the United Arab Emirates and Yemen. It includes the neutral zone between Saudi Arabia and Iraq.

OECD Europe

Comprises: Austria, Belgium, the Czech Republic, Denmark, Finland, France, Germany, Greece,Hungary, Iceland, Ireland, Italy, Luxembourg, the Netherlands, Norway, Poland, Portugal, Spain,Sweden, Switzerland, Turkey and the United Kingdom.

Organization of Petroleum Exporting Countries (OPEC)

Comprises: Algeria, Indonesia, Iran, Iraq, Kuwait, Libya, Nigeria, Qatar, Saudi Arabia, the UnitedArab Emirates and Venezuela.

Other Developing Asia

Comprises: Afghanistan, Bangladesh, Bhutan, Brunei, Chinese Taipei, Fiji, French Polynesia, Indonesia,Kiribati, Democratic People’s Republic of Korea, Malaysia, Maldives, Mongolia, Myanmar, Nepal,New Caledonia, Pakistan, Papua New Guinea, the Philippines, Samoa, Singapore, Solomon Islands,Sri Lanka, Thailand, Vietnam and Vanuatu.

Western Europe

Comprises: Austria, Belgium, the Czech Republic, Denmark, Finland, France, Germany, Greece,Hungary, Iceland, Ireland, Italy, Luxembourg, the Netherlands, Norway, Portugal, Spain, Sweden,Switzerland, Turkey and the United Kingdom.

227-234 Annexe 5 17/11/2004 16:42

ABBREVIATIONS AND ACRONYMS

AFR Africa

API American Petroleum Institute

ASU Air Separation Unit

AUS Australia and New Zealand

BKB Brown Coal Briquettes

CA Chemical absorption

CaCO3 Calcium carbonate

CAN Canada

CaO Calcium oxide

CAT Carbon Abatement Technologies

CC Combined cycle

CCC Clean Coal Centre

CCS CO2 Capture and Storage

CDM Clean Development Mechanism

CENS CO2 for EOR in the North Sea

CERT Committee on Energy Research and Technology

CFB Circulating Fluid Bed

CHI China

CHP Combined Heat and Power

CO2 Carbon dioxide

CRUST CO2 Re-use through Underground Storage

CSA Central and South America

CSLF Carbon Sequestration Leadership Forum

CUCBM China United Coal-bed Methane Corporation

DME Dimethyl Ether

DOE Department of Energy

DRI Direct Reduced Iron

ECBM Enhanced Coal-bed Methane recovery


227-234 Annexe 5 17/11/2004 16:42


EEU Eastern Europe

EGR Enhanced Gas Recovery

EOH Ethanol

EOR Enhanced Oil Recovery

EPR European Pressurized water Reactor

ESPOO ECE convention on Trans-boundary Impact Assessment

ETP Energy Technology Perspectives

ETS EU Emissions Trading Scheme

ETSAP Energy Technology Systems Analysis Programme

EU European Union

EUR Euro

FCC Fluid Catalytic Cracker

FGD Flue Gas Desulphurization

FSU Former Soviet Union

FT Fischer-Tropsch

GB Governing Board

GDP Gross Domestic Product

GHG Greenhouse Gas

GIS Geographical Information System

GTL Gas-to-Liquids

HTGR High Temperature Gas-cooled Reactor

IEA International Energy Agency

IET International Emissions Trading

IGCC Integrated Gasification Combined Cycle

IND India

IPCC Intergovernmental Panel on Climate Change

JI Joint Implementation

JPN Japan

LHV Lower Heating Value

LNG Liquefied Natural Gas

227-234 Annexe 5 17/11/2004 16:42

LPG Liquefied Petroleum Gas

LTF Low Temperature Flash

MEA Middle East

MEA MonoEthanol Amine

MeOH Methanol

MEX Mexico

MgCl2 Magnesium Chloride

MgO Magnesium Oxide

NGO Non-Governmental Organisation

NOx Nitrogen oxides

ODA Other Developing Asia

OECD Organisation for Economic Co-operation and Development

OPEC Organisation of Petroleum Exporting Countries

OSPAR Oslo Convention and Paris Convention for the Protection of the Marine Environment of the North-East Atlantic

OxF OxyFueling

PA Physical Absorption

PFBC Pressurized Fluidized Bed Combustion

PM10 Particulate Matter of less than 10 micron diameter

PPP Purchasing Power Parity

PV PhotoVoltaics

RD&D Research, Development and Demonstration

SACS Saline Aquifer CO2 storage

SC Supercritical

SCSC Supercritical steam cycle

SKO South Korea

SOFC Solid Oxide Fuel Cells

SO2 Sulphur dioxide

TPES Total Primary Energy Supply

UNCLOS United Nations Convention for the Law of the Sea


227-234 Annexe 5 17/11/2004 16:42

UNFCCC United Nations Framework Convention on Climate Change

USA United States of America

USC Ultra Supercritical

USCSC Ultra Supercritical steam cycle

USD United States Dollars

WEO World Energy Outlook

WEU Western Europe

UNITS

MJ megajoule = 106 joules

GJ gigajoule = 109 joules

PJ petajoule = 1015 joules

EJ exajoule = 1018 joules

t tonne = metric ton = 1000 kilogrammes

Mt megatonne = 103 tonnes

Gt gigatonne = 109 tonnes

kW kilowatt = 103 watts

MW megawatt = 106 watts

GW gigawatt = 109 watts

TW terawatt = 1012 watts

bbl (blue) barrel

BOE Barrels of Oil Equivalent. 1 BOE = 41.868 GJ

°C degrees Celsius

kWh kilowatt-hour

mD millidarcies = 10-3 darcies

mils 0.001 US dollar

MPa megapascal = 106 Pa

Nm3 Normal cubic metre. Measured at 0 degrees Celsius and a pressure of 1.013 bar.

ppm parts per million

Pa pascal

Wp watts peak


227-234 Annexe 5 17/11/2004 16:42

REFERENCES

Abdul Baqi, M. and N.G. Saleri (2004), “Fifty-year crude oil scenarios: Saudi Aramco’s perspective”, Center for Strategic and International Studies, Washington DC, 24 February,www.saudiaramco.com/sa/webServer/general/Summary_Fifty_Year_Crude_Oil_Supply.pdf.

AD700 (not dated), “Advanced 700 °C PF power plant”, http://ad700.techwise.dk/.

Air Products (2002), “Market outlook for dimethyl ether (DME)”, Topical report, www.netl.doe.gov/cctc/resources/pdfs/estmn/DME2.Top.pdf.

American Petroleum Institute (2002), “Emissions estimating and reporting”, paper presented at the 2nd API Conference on Voluntary actions by the oil and gas industry to addressclimate change, Houston, Texas, USA, 20-21 November.

Armor, A.F. (1996), “Fossil power plant residual life optimization in the USA”, Modern Power Systems, Vol. 16, No. 1, pp. 41, 43-45.

Audus, H. (2003), “The Sleipner and Snøhvit CO2-injection projects”, paper presented at the firstworkshop of the Canadian CO2 capture and storage technology roadmap, Calgary, AB,Canada, 18-19 September.

Bachu, S. (2000), “Sequestration of CO2 in geological media: criteria and approach for site selection in response to climate change”, Energy Conversion and Management, Vol. 41,No. 9, pp. 953-970.

Bachu, S. (2003), “Screening and ranking of sedimentary basins for sequestration of CO2 in geological media in response to climate change”, Environmental Geology 44, pp. 277-289.

Bachu, S. and J.J. Adams (2003), “Sequestration of CO2 in geological media in response to climate change: capacity of deep saline aquifers to sequester CO2 in solution”, EnergyConversion and Management, Vol. 44, No. 20, pp. 3151-3175.

Bachu, S., J.J. Adams, K. Michael and B.E. Buschkuehle (2003), “Acid gas injection in the Alberta basin: a Commercial-Scale Analogue for CO2 Geological Sequestration in SedimentaryBasins”, Alberta Geological Survey, Edmonton, Canada.

Bataille, C. and C. Birraux, (2003), “Lifetime of nuclear power plants and new designs of reactors”, French Parliamentary Office for Scientific and Technological Assessment, Paris,France.

Benson, S. (2004), “Status and current issues in geologic storage of carbon dioxide”, presentation at the 7th International Conference on Greenhouse Gas Control Technologies, 5-9 September, Vancouver, Canada.

Benson, S. (2004), presentation for the CSLF Ministerial Meeting, Melbourne, September,www.cslf.org.

Berge, G. (2003), “CO2 – fra problem til ressurs.”, Miljøforum, Norwegian Oil Industry Association, Stavanger, 25 Norway.

REFERENCES 235

235-250 References 18/11/2004 17:30


Bergen, F. van, H.J.M. Pagnier, P. David and B.M. Kross (2000), “Inventory of the potential for enhanced coalbed methane production with carbon dioxide disposal”, TNO, Utrecht, TheNetherlands, www.ergweb.com/methane/pdf/ban_bergin.pdf.

Bergen, F. van, K. Damen, J.S. Ribberink, H.J.M. Pagnier and A. Faaij (2003), “Feasibility study on CO2 sequestration and enhanced CBM production in Zuid-Limburg”, Netherlands Agency forEnergy and the Environment (NOVEM), Utrecht, The Netherlands.

Bergen, F. van, J. Gale, K.J. Damen, and A.F.B. Wildenborg, (2004), “Worldwide selection of early opportunities for CO2-enhanced oil recovery and CO2-enhanced coal bed methaneproduction”, Energy 29, pp. 1611-1621.

Bertani, R. (2003), “What is geothermal potential”, IGA News No. 53, http://iga.igg.cnr.it/IGA/potential.pdf.

B.H.P. Billiton (2003), “Opportunities for Australian CBM companies in China”, www.bhpbilliton.com/bbContentRepository/Presentations/ChinaCBMpresentation4Dec2003.pdf.

Bolland, O. and H. Undrum, (2003), “A novel methodology for comparing CO2 capture options for natural gas-fired combined cycle plants”, Advances in Environmental Research, Vol. 7, No. 4, pp. 901-911.

Brandvoll, O. and O. Bolland, (2002), “Inherent CO2 capture using chemical looping combustion in a natural gas-fired power cycle”, paper presented at ASME TURBO EXPO 2002: land, seaand air, Amsterdam, The Netherlands, 3-6 June. Paper GT-2002-30129, New York, NY, USA,ASME International.

Breton, D.L. and P. Amick, (2002), “Comparative IGCC cost and performance for domestic coals”, paper presented at 2002 gasification technologies conference, 27-30 October, SanFrancisco, CA, USA. Arlington, VA, USA, Gasification Technologies Council.

Bruant, R.G., M.A. Celia, A.J. Guswa, and C.A. Peters (2002), “Safe storage of CO2 in deep saline Aquifers”, Environmental Science and Technology, Vol. 36, No. 11, pp. 240A-245A.

C3 Views. (2003), “Sparking a less carbon-intensive future – greenhouse gas technologies: enhanced coalbed methane and biomass power”, C3 Views, No. 5, pp. 1-12,www.climatechangecentral.com/resources/c3views/C3Views200301_issue5.pdf .

Caldeira, K. (2002), “Monitoring of ocean storage projects”, Proceedings of the IPCC workshop on Carbon Dioxide Capture and Storage, Regina, Canada, 18-21 November. Energy ResearchCentre of the Netherlands, Petten, http://www.climatepolicy.info/ipcc/ipcc-ccs-2002.

Caldwell, J.C. (2001), “Testimony for the United States House of Representatives Committee on Resources”, Subcommittee on Energy and Mineral Resources,www.dnr.state.la.us/SEC/EXECDIV/TECHASMT/data/oil_gas/House_OCS_051401.pdf.

Canadian Clean Power Coalition (2004), “CCPC Phase 1 Executive Summary”, paper presented at the GHGT7 conference, Vancouver, 5-9 September.

Capra, M. (2004), “Report on ZETs related R&D in Italy”, presentation at the 46th meeting of the IEA Fossil Fuel Working Party, 8th-9th July, Paris.

235-250 References 18/11/2004 17:30

Castle, W.F. (2002), “Air separation and liquefaction: recent development and prospects for the beginning of the new millennium”, International Journal of Refrigeration, Vol. 25, No. 1,pp. 158-172.

Celia, M.A. (2003), “Environmental considerations and leakage estimation for CO2 injection in mature sedimentary basins”, paper presented at the workshop on reservoir storage of CO2,Bergen, Norway, 6-7 November, http://www.mi.uib.no/CO2/.

Celia, M.A. and S. Bachu (2003), “Geological sequestration of carbon dioxide: is leakage unavoidable and acceptable?”, Proceedings of the 6th International Conference onGreenhouse Gas Control Technologies, Kyoto, Japan, 30 September - 4 October 2002. Oxford,UK, Elsevier Science (Pergamon), pp. 161-166.

Chapel, D., C.L. Mariz, and J. Ernest (1999), “Recovery of CO2 from flue gases: commercial trends”, paper presented at the Canadian Society of Chemical Engineers annual meeting,Saskatoon, SK, Canada, 4-6 October.

Chemical Market Reporter (2004), “GTL could become major chemicals feedstock”, 12 January.

Christensen, N.P. (2003), “CO2 capture and geological storage”, paper presented at the international energy conference, Risø, Denmark, 19-21 May.

Clarke, S.C. (2003), “CO2 management – a refiners perspective”, www.nrcan.gc.ca/es/etb/cetc/combustion/co2trm/htmldocs/technical_reports_e.html.

Clean Power Coalition (2004), http://www.canadiancleanpowercoalition.com/.

Clemens, T.G. and K. Wit, (2001), “Zero emission power generation. Power plant concepts and CO2 injection into gas fields”, Report EP-2001-5403, Rijswijk, Shell Technology EP, theNetherlands.

CO2CRC (2004), “Carbon dioxide capture and storage”, Research development & demonstration in Australia. A technology roadmap, Cooperative Research Centre for Greenhouse GasTechnologies, Canberra, ACT, Australia.

Coal Utilization Research Council (2002), “Technology platforms: specific performance parameters and milestones through 2020”, www.coal.org/content/roadmap.htm.

Cody, G.D. and T. Tiedje (1997), “A learning curve approach to projecting cost and performance for photovoltaic technologies”, proceedings of SPIE, Vol. 3138, pp. 78-89.

Coleman (2004), “CO2 for EOR in the North Sea (CENS) – Update on project status”, Sharp IOR Newsletter, http://ior.rml.co.uk/issue7/articles/kindermorgan/.

Cooretec (2003), “Forschungs- und Entwicklungskonzept für emissionsarme fossil befeuerte Kraftwerke. Bericht der COORETEC-Arbeitsgruppen”, Bundesministerium für Wirtschaft undArbeit’, dokumentation Nr. 527. ISSN 0342-9288, Berlin.

Craig, P. (2003), “Carbon sequestration – A Sierra club view”, paper presented at the 2nd annual conference on carbon sequestration, Alexandria, VA, USA, 5-8 May.

Curry, T., D.M. Reiner, S. Ansolabehere, and H. Herzog (2004), “How aware is the public of carbon capture and storage?”, paper presented at the 7th International Conference onGreenhouse Gas Control Technologies, 5-9 September, Vancouver, Canada.

REFERENCES 237

235-250 References 18/11/2004 17:30


Czisch (2003), “Daten zu regenerative energien”(sic), www.iset.uni-kassel.de/abt/w3-w/folien/Allgemein/Daten.html.

David, J. and H. Herzog (2001), “The cost of carbon capture”, Procedings of the 5th International Conference on Greenhouse Gas Control Technologies, Cairns, Qld., Australia. 13-16 August 2000, Collingwood, Vic., Australia, CSIRO Publishing, pp. 985-990.

Dickson, B., I. Yashayaev, J. Meincke, B. Turrell, S. Dye, and J. Holfort (2002), “Rapid freshening of the deep North Atlantic Ocean over the past four decades”, Nature, Vol. 416, No. 6883,pp. 832-837.

Dijkstra, J.W. and D. Jansen, (2003), “Novel concepts for CO2 capture with SOFC”, Proceedings of the 6th International Conference on Greenhouse Gas Control Technologies, Kyoto, Japan. 1-4 October 2002. Oxford, UK, Elsevier Science (Pergamon), pp. 161-166.

Department of Energy, United States (2004), “FutureGen. Integrated hydrogen, electric power production and carbon sequestration research initiative”, Office of Fossil Energy, March.

DTI (2003), “Review of the feasibility of carbon dioxide capture and storage in the UK”, DTI/Pub URN 03/1261, Department of Trade and Industry, London.

DTI (2004), “A carbon abatement technologies strategy for fossil fuel power generation”, consultation document, Department of Trade and Industry, London,www.dti.gov.uk/energy/coal/cfft/consdoc.pdf.

ECMWF (2003), “Wind and solar data originally from the ERA-15 global re-analysis project1979-1993”, European Centre for Medium-Range Weather Forecasts,http://www.ecmwf.int/research/era/ERA-15/index.html.

Economist (2004), “A very big French turn-off”, Special report on Electricité de France, The Economist, pp. 65-67, 3 July.

Elvestad, O. (2003), “Case study: CO2 capture and storage from a 400 MW gas fired power plant at Kårstø, Norway”, paper presented at the IEA CO2 capture workshop, Paris, France,24 January.

ERI (2003), “Comprehensive report on China’s sustainable energy development and carbon emission scenario analysis”, Energy Research Institute of the State Development andPlanning Commission, Beijing, China.

ENI (2004), “World oil and gas review 2004”, Rome, Italy, www.agip.it/

Espinoza, R.L., A.P. Steynberg, B. Jager, and A.C. Vosloo (1999), “Low temperature Fischer-Tropsch synthesis from a Sasol perspective”, Applied Catalysis A: General 186, pp. 13-26.

Esposito, P.R. and C.D. Locke (2004), “Education & Outreach Programs: Important factors in sequestration’s future”, paper presented at the 2nd Annual Conference on CarbonSequestration, Alexandria, VA, USA, 5-8 May.

ETSAP (2003), IEA Energy Technology Systems Analysis Programme: www.etsap.org/.

Eurocoal (2003), Coal industry across Europe, Eurocoal, Brussels, Belgium.

235-250 References 18/11/2004 17:30

European Commission (2004), “European hydrogen and fuel cell technology platform”, meeting of the MS mirror group, Brussels, 22 June 2004,http://www.fuelcells.ro/portal/ro/publishing/2/ms_mirror_group_-_aps.pdf.

Ewers, J., J. Engelhard, I. Strömberg, H. Altmann and W. Bledau (2003), “The development of coal-fired power plant technology in the light of preventive climate protection”, paperpresented at RWE Rheinbraun/Vattenfall Europe generation meeting, Cologne/Berlin,Germany, 9 April.

Farla, J.C.M., C.A. Hendriks and K. Blok (1995), “Carbon dioxide recovery from industrial processes”, Climatic Change, Vol. 29, No. 4, pp. 439-461.

Fishbone, L.G. and H. Abilock (1981), “MARKAL, a linear-programming model for energy systems analysis: technical description of the BNL version”. Energy Research, Vol. 5, pp. 353-375.

Fleming, A. (2002), “New materials for high temperature service in power generation”, Operation Maintenance and Materials Issue, Vol. 1, No. 2, www.ommi.co.uk/.

Fleming, A. and D. Foster (2001), “Ageing of coal-fired power plants”, Report no. CCC/34, IEA Coal Research, London, UK.

Freund, P. (2003), “Current assessment and future prospects”, paper presented at the seminar on CO2 underground sequestration: the green way forward, Institut Français du Pétrole,Rueil-Malmaison, France, 2 October.

Freund, P. and J. Davison. (2002), “General overview of costs”, in the Proceedings of the IPCC workshop on Carbon Dioxide Capture and Storage, Regina, Canada, 18-21 November. Energyresearch Centre of the Netherlands, Petten.

Friedmann, S.J. and D. Nummedal (2003), “Reassessing the geological risk of seal failure for saline aquifers and EOR projects”, paper presented at the 2nd Annual Carbon SequestrationConference, Alexandria, VA, USA, 5-8 May.

Gale, J. (2002), “Overview of CO2 emission sources, potential, transport and geographical distribution of storage possibilities”, in the Proceedings of the IPCC workshop on CarbonDioxide Capture and Storage, Regina, Canada, 18-21 November. Energy Research Centre ofthe Netherlands, Petten.

Gale, J. (2004), “Using coal seams for CO2 sequestration”, Geologica Belgica, Vol. 7, No. 3/4, in press.

Geipel, H. (2004), Congress “Innovative technologies for generating electricity – towards zero 46th emission coal fired and gas power plants, Berlin, 10-12 May 2004”, presentation at themeeting of the Fossil Fuel Working Party, 8th-9th July, 2004, IEA, Paris, France.

GFZ Potsdam (2004), “In-situ R&D laboratory for geological storage of CO2 - CO2SINK”, Internet: www.gfz-potsdam.de/pb5/pb51/projects/project-CO2SINK/content_en.html.

Gielen, D.J. (2003), “CO2 removal in the iron and steel industry”. Energy Conversion and Management, Vol. 44, No. 7, pp. 1027-1037.

Gielen, D.J. and S. Karbuz (2003), “Challenges in energy and environment modelling: a materials perspective” IEA, Paris, EET/2003/05, http://library.iea.org/dbtw-wpd/textbase/papers/2003/eet05.pdf

REFERENCES 239

235-250 References 18/11/2004 17:30


Gielen and Moriguchi (2003), “CO2 in the iron and steel industry: an analysis of Japanese emission reduction potentials”, Energy Policy 30, pp. 849-863.

Gielen, D.J., F. Unander, N. Mattsson and R. Sellers (2004), “Technology learning in the ETP model”, paper presented at the 6th IAEE conference ‘Modelling of energy economics andpolicy’, Zürich, Switzerland, 2-3 September.

Goerne, G. von (2004), presentation for the CSLF Ministerial Meeting, Melbourne 2004.

Goldberg, P., Z.-Y. Chen, W. Connor, R. Walters and H. Ziock, (2001), “CO2 mineral sequestration studies in the US”, paper presented at the First National Conference on CarbonSequestration, Washington, DC, USA, 14-17 May, Internet:www.netl.doe.gov/publications/proceedings/01/carbon_seq/6c1.pdf.

Gorgon (2004), “Gorgon Australian Gas Project”, www.gorgon.com.au/.

Gray, D. and G. Tomlinson (2001), “Coproduction of ultra clean transportation fuels, hydrogen, and electric power from coal”, Mitretek technical report No. MTR 2001-43,www.angtl.com.pdfs/IGCCGTL.pdf.

Green, D.W. and G.P. Willhite (1998), “Enhanced oil recovery”, Textbook Vol. 6, Society of Petroleum Engineers Richardson, TX, USA.

Gunter, W.D., R.J. Chalaturnyk and J.D. Scott (2001), “Monitoring of aquifer disposal of CO2: experience from underground gas storage and enhanced oil recovery”, paper presented atNETL workshop on carbon sequestration science, Pittsburgh, PA, USA, 22-24 May,www.netl.doe.gov/publications/proceedings/01/carbon_seq_wksp/1bGeologic.PDF.

Gupta, N., B. Sass and J. Sminchak (2002), “Feasibility of CO2 sequestration in deep saline reservoirs in the Midwestern USA”, paper presented at the North-Central and SoutheasternSections of the Geological Society of America meeting, Lexington, KY, USA, 4 April,www.midcarb.org/reports.shtml.

Gupta, M., I. Coyl, and K. Thambimuthu, (2003), “CO2 capture technologies and opportunities in Canada”, paper presented at the 1st workshop of the Canadian CO2 capture and storagetechnology roadmap, Calgary, AB, Canada, 18-19 September.

Haefeli, S., M. Bosi and C. Philibert (2004), “Carbon Dioxide Capture and Storage Issues – Accounting and Baselines under the United Framework Convention on Climate Change (UNFCCC)”, IEA Information Paper, IEA/OECD, Paris, France.

Harmon, C. (2000), “Experience curves of photovoltaic technology”, IIASA Interim Report IR-00-014. , Laxenburg, Austria, International Institute for Applied Systems Analysis,http://www.iiasa.ac.at/Publications/Documents/IR-00-014.pdf.

Hawkins, D.G. (2004), “CCS: Key questions for us.”, presentation at CAN Europe meeting, 27 May, http://www.climnet.org/CTAP/workshop2004/PresentationsWS2.htm.

Henderson, C. (2003), “Clean coal technologies roadmaps”, Report no. CCC/75, IEA Clean Coal Centre, London, UK.

Hendriks, C.A., E. Worrell, L. Price, N. Martin, O. Meida, D. de Jager and P. Riemer (1999)“Emission reduction of greenhouse gases from the cement industry”, Proceedings of the

235-250 References 18/11/2004 17:30

4th International Conference on Greenhouse Gas Control Technologies, Interlaken, Switzerland,30 August–2 September 1998, Elsevier Science (Pergamon), Oxford, UK, pp. 939-944.

Hendriks, C.A., A.F.B. Wildenborg, K. Blok, F. Floris and J. van Wees (2001), “Costs of carbon dioxide removal by underground storage”, Proceedings of the 5th International Conference onGreenhouse Gas Control Technologies, Cairns, Qld., Australia. 13-16 August 2000,Collingwood, Vic., Australia, CSIRO Publishing, pp. 967-972.

Hepple, R.P. and S.M. Benson (2003), “Implications of surface seepage on the effectiveness of geological storage of carbon dioxide as a climate change mitigation strategy”, Proceedingsof the 6th International Conference on Greenhouse Gas Control Technologies, Kyoto, Japan, 1-4 October 2002, Oxford, UK, Elsevier Science (Pergamon), pp. 261-266.

Herzog, H. (2002), “Carbon sequestration via mineral carbonation: Overview and assessment”, MIT, Boston.

Herzog, H. (2003), “Pathways for reducing the cost of CO2 capture”, paper presented at the CO2 capture workshop, IEA, Paris, France, 24 January.

Holt, T., J.I. Jensen and E. Lindeberg (1995), “Underground storage of CO2 in aquifers and oil reservoirs”, Energy Convers. Mgmt, Vol. 36, No. 6-9, pp. 535-538.

Hustad, C.W. (2003), “Infrastructure for CO2 collection, transport and sequestration”, paper presented at the 3rd Nordic mini-symposium on carbon dioxide storage, Trondheim, Norway,3 October.

IEA (2000), “Experience curves for energy technology policy”, IEA/OECD, Paris, France.

IEA (2001), “World Energy Outlook Insights. Assessing Today’s Supplies to Fuel Tomorrow’s Growth”, IEA/OECD, Paris, France.

IEA (2002a), “World Energy Outlook 2002”, IEA/OECD, Paris, France.

IEA (2002b), “CO2 emissions from fuel combustion 1971-2000”, IEA/OECD, Paris, France.

IEA (2003a), “Renewables for power generation”, IEA/OECD, Paris, France.

IEA (2003b), “Cool appliances. Policy strategies for energy-efficient homes”, IEA/OECD, Paris, France.

IEA (2003c), “World Energy Investment Outlook – 2003 insights”, IEA/OECD, Paris, France.

IEA (2004a), “World Energy Outlook. 2004 Insights”, IEA/OECD, Paris, France.

IEA (2004b), “30 Years of Energy Use in IEA Countries”, IEA/OECD, Paris, France.

IEA (2004c), “Legal Aspects of Storing Carbon Dioxide”, IEA, Paris, France.

IEA GHG (1998), “Enhanced coal bed methane recovery with CO2 sequestration”, IEA Greenhouse Gas R&D Programme, Cheltenham, UK.

IEA GHG (1999), “Putting carbon back into the ground”. IEA Greenhouse Gas R&D Programme, Cheltenham, UK.

IEA GHG (2000), “Leading options for the capture of CO2 emissions at power stations”. IEAGreenhouse Gas R&D Programme, Cheltenham, UK.

REFERENCES 241

235-250 References 18/11/2004 17:30


IEA GHG (2002a), “Transmission of CO2 and energy”, report no. PH4/6, March. IEA Greenhouse Gas R&D Programme, Cheltenham, UK.

IEA GHG (2002b), “Information brochure ‘Ocean Storage of CO2’”, IEA Greenhouse Gas R&D Programme, Cheltenham, UK.

IEA GHG (2003), “Potential for improvement in gasification combined cycle power generation with CO2 capture”. IEA Greenhouse Gas R&D Programme, Cheltenham, UK.

IEA GHG (2004), R&D project database, www.co2sequestration.info.

IEA GHG (2004b), “Ship transport of CO2”. IEA Greenhouse Gas R&D Programme, Cheltenham, UK.

Imperial Oil (2000), “Annual report pursuant to section 13 or 15(d) of the securities exchange act of 1934”, www.imperialoil.com/pdf/200010k.pdf.

IPCC (2001), “Climate change 2001: emissions, concentrations and temperature changescorresponding to different stabilization levels for CO2 concentrations”,www.grida.no/climate/ipcc_tar/slides/large/02.18.jpg.

Jones, B. (2004), “Research into zero emissions technologies in Australia”, presentation at the 46th meeting of the Fossil Fuel Working Party, 8th-9th July, 2004, IEA, Paris, France.

Jordal, K., M. Anheden, J. Yan, and L. Strömberg, (2004), “Oxyfuel combustion for coal-fired power generation with CO2 capture – opportunities and challenges”, paper presented at the7th International Conference on Greenhouse Gas Control Technologies, 5-9 September,Vancouver.

Junginger, M., A. Faaij, and W.C. Turkenburg (2005), “Global experience curves for wind farms”, Energy Policy, vol. 33, no. 2, pp. 133-150.

Kaarstad, O. (2003), “Business development opportunities”, paper presented at the IPIECA workshop on carbon dioxide capture and geological storage, 21-22 October, Brussels,Belgium.

Kaldis, S.P., G. Skodas and G.P. Sakellaropoulos (2003), “Energy and capital cost analysis of CO2

capture in coal IGCC processes via gas separation membranes”. Fuel Processing Technology,Vol. 85, No. 5, pp. 337-346.

Keith, D. and S. Pacala (2004), “Timescales for CO2 storage: How long is long enough?”, paper presented at the 7th International Conference on Greenhouse Gas Control Technologies, 5-9 September, Vancouver.

Keith, D. (2004), Private communication.

Kenny, A. (2004), “Electricity and hydrogen production from the nuclear pebble bed modular reactor”, paper presented at the International Energy Workshop (IEW), 22-24 June, Paris,France, www.iiasa.ac.at/Research/ECS/IEW2004/docs/Kenny_2004IEW.ppt.

Kinder Morgan (2002), “Rules of thumb”, www.kne.com/co2/flood.cfm.

235-250 References 18/11/2004 17:30

Klara, S. (2003), “Systems analysis supporting the carbon sequestration technology roadmap”, paper presented at the 2nd Annual Conference on Carbon Sequestration, 5-8 May,Alexandria, VA, USA.

Kuik, O. and R. Gerlagh (2003), “Trade liberalization and carbon leakage”, The Energy Journal, Vol. 24, No. 3, pp. 97-120.

Lackner, K.S., D.L. Butt, and C.H. Wendt (1997), “Progress on binding CO2 in mineral substrates”, Energy Conversion and Management, Vol. 38, supplement, pp. S259-S265.

Laherrere, J. (2004), “Natural gas supply future”, paper presented at the International EnergyWorkshop (IEW), 22-24 June, Paris, France,www.iiasa.ac.at/Research/ECS/IEW2004/docs/Laherrere_2004IEW.pdf.

Lange, T. de, M. Beeldman, J.Kiel, H. den Uil and J. Veenkamp (2001), “Co-production of fuels as an option for Demkolec?”, report no. ECN-C-01-004, Energy Research Centre of theNetherlands (ECN), Petten, The Netherlands.

Larson, E.D. (2002), “The Princeton-Tsinghua collaboration on low emission energy technologies and strategies for China”, paper presented at the Carbon Mitigation Initiative hydrogenmeeting, 16 January, Princeton, NJ, USA,http://www.princeton.edu/~energy/presentations/pdfs/Larson.pdf.

Lindeberg, E. (1997), “Escape of CO2 from aquifers”, Energy Conversion and Management, Vol. 38, supplement, pp. S235-S240.

Lippman, M.J. and S.M. Benson (2002), “Relevance of underground natural gas storage to geologic sequestration of carbon dioxide”, Berkeley, CA, USA, Lawrence Berkeley NationalLaboratory.

Loulou, R., G. Goldstein and K. Noble (2004) Documentation for the MARKAL family of models. IEA Energy Technology Systems Analysis Programme. Internet: www.etsap.org

Marchetti, C. (1977), “On geoengineering and the CO2 problem”, Climatic Change, Vol. 1, No. 1, pp. 59-68.

Mariott, J. (2000), “Implementation of the Sasol slurry phase distillate process”, paper presented at the 16th world petroleum congress, 11-15 June, Calgary, AB, Canada.

Markewitz, P., D. Martinsen and S. Voegele (2004), “The future role of CO2 capture as part of a German mitigation strategy”, paper presented at the 7th International Conference onGreenhouse Gas Control Technologies, 5-9 September, Vancouver, Canada.

Marsh, G. (2003), “Carbon dioxide capture and storage: a win-win option? (The economic case)”, report no. COAL R233, DTI/PUB URN 03/812, Department of Trade and Industry, London,UK.

Marsh, G., J. Bates, H. Haydock, N. Hill, C. Clark, and P. Freund (2003), “Application of CO2

removal to the Fischer-Tropsch process to produce transport fuel”. Proceedings of the 6th

International Conference on Greenhouse Gas Control Technologies, 1-4 October 2002, Kyoto,Japan. Oxford, UK, Elsevier Science (Pergamon), pp. 45-50.

REFERENCES 243

235-250 References 18/11/2004 17:30


Mathieu, P. (2003), “IPCC special report on CO2 capture and storage”, paper presented at the climate technologies assessment workshop, 23-24 September, Brussels, Belgium,www.climnet.org/CTAP/workshop2003/1-2mathieu-capture.pdf.

Matthiassen, O.M. (2003), “CO2 as injection gas for enhanced oil recovery and estimation of the potential on the Norwegian continental shelf”, Norwegian University of Science andTechnology (NTNU), Department of Petroleum Engineering and Applied Geophysics,Trondheim/Stavanger.

McKee, B. (2002), “Solutions for the 21st century”. Working Party on Fossil Fuels, IEA, Paris, France.

McKee, B. (2004a), “Carbon sequestration leadership forum. Final draft technology roadmap”, www.cslforum.org.

McKee, B. (2004b), “Zero Emissions Program Activities in the United States”, paper presented at the 46th meeting of the Fossil Fuel Working Party, 8th-9th July, 2004, IEA, Paris, France.

Mimura, T., T. Nojo, M. Iijima and S. Mitsuoka (2002), “Development and application of flue gas carbon dioxide recovery technology”, paper presented at the 7th meeting of the IEA AdvisoryGroup on Oil and Gas Technology, 15 November, Stavanger, Norway.

MIT (2003), “The future of nuclear power”, Massachusetts Institute of Technology, Cambridge, MA, USA, www.mit.edu/afs/athena/org/n/nuclearpower/.

Möllersten, K., J. Yan and J.R. Moreira (2003), “Potential market niches for biomass energy with CO2 capture and storage – opportunities for energy supply with negative CO2 emissions”,Biomass and Bioenergy, Vol. 25, No. 3, pp. 273-285.

Möllersten, K., L. Gao, J. Yan and M. Obersteiner (2004), “Efficient energy systems with CO2

capture and storage from renewable biomass in pulp and paper mills”, Renewable Energy,Vol. 29, No. 9, pp. 1583-1598.

Moritis, G. (2004), “EOR continues to unlock oil resources”, Oil & Gas Journal, April 12, p. 53-65.

Morris, B. (2004), “UK country report”, presentation at the 46th meeting of the Fossil Fuel Working Party, 8th-9th July, 2004, IEA, Paris, France.

Nakicenovic, N., J. Alcamo, G. Davis, H. De Vries, J. Fenhann, S. Gaffin, K. Gregory, A. Grübler, T. Jung, T. Kram, E. La Rovere, L. Michaelis, S. Mori, T. Morita, W. Pepper, H. Pitcher, L. Price,K. Riahi, A. Roehrl, H.H. Rogner, A. Sankovski, M. Schlesinger, P. Shukla, S. Smith, R. Swart, S. van Rooijen, N. Victor and Z. Dadi (2000), “Special Report on Emissions Scenarios”,Cambridge University Press, UK.

Neij, L. (1997), “Use of experience curves to analyse the prospects for diffusion and adoption of renewable energy technology”, Energy Policy, Vol. 25, No. 13, pp. 1099-1107.

Norwegian Petroleum Directorate (2002), “Presentation for the 7th meeting of the IEA Advisory Group on Oil and Gas Technology”, 15 November, Stavanger, Norway, www.iea.org/dbtw-wpd/textbase/work/2002/stavanger/norsk_hydro.pdf.

NRCAN(2004), “Canada”, Natural Resources Canada, presentation at the 46th meeting of the Fossil Fuel Working Party, 8th-9th July, IEA, Paris, France.

235-250 References 18/11/2004 17:30

Nsakala, N., G. Liljedahl, J. Marion, C. Bozzuto, H. Andrus and R. Chamberland, R. (2003), “Greenhouse gas emissions control by oxygen firing in circulating fluidized bed boilers”,paper presented at the 2nd Annual Conference on Carbon Sequestration, 5-8 May,Alexandria, VA, USA.

NYU Stern (2002), “Country default spreads and risk premiums”, www.stern.nyu.edu/~adamodar/New_Home_Page/datafile/ctryprem.html.

Ogawa, T., N. Inoue, T. Shikada and Y. Ohno (2003), “Direct dimethyl ether synthesis”. Journal of Natural Gas Chemistry, Vol. 12, No. 4, pp. 219-227.

Oil & Gas Journal Energy Database (2001), “Historical refinery survey”,http://orc.pennnet.com/energydatabase/energydatabase.cfm.

Oldenburg, C.M. and S.M. Benson (2001), “Carbon sequestration with enhanced gas recovery: identifying candidate sites for pilot study”, paper presented at the 1st Annual Conference onCarbon Sequestration, 14-17 May, Washington, DC, USA,www.netl.doe.gov/publications/proceedings/01/carbon_seq/2a4.pdf.

Otter, N. (2004) “Reduce greenhouse global signature – role of carbon capture and storage”, Presentation for the CSLF, Melbourne, September 2004, www.cslf.org.

Over, J., J. de Vries, and J. Stork, (1999), “Removal of CO2 by storage in the deep underground, chemical utilization and biofixation”, Netherlands Agency for Energy and the Environment (NOVEM), Utrecht, The Netherlands.

Pacala, S.W. (2003), “Global constraints on reservoir leakage”, Proceedings of the 6th

International Conference on Greenhouse Gas Control Technologies, 1-4 October 2002, Kyoto,Japan. Oxford, UK, Elsevier Science (Pergamon), pp. 267-272.

Paez, R., L. Luzardo, and J. Guitian (2002), “Current and future upgrading options for the Orinoco heavy crude oils”, paper presented at the 16th World Petroleum Congress, 11-15 June, Calgary, AB, Canada.

Paterson, I.R. and J.D. Wilson (2002), “Use of damage monitoring systems for component life optimisation in power plant”, International Journal of Pressure Vessels and Piping, Vol. 79,No. 8, pp. 541-547.

Pehnt, M. (2004), “Fuel cells in the power market: Separating the hope from the hype”, presentation at CAN Europe meeting, 27 May, Brussels, Belgium.www.climnet.org/CTAP/workshop2004/PresentationsWS2.htm.

Pollack, H.N., S.J. Hurter and J.R. Johnson (1991), “A new global heat flow compilation”, University of Michigan, Department of Geological Sciences, Ann Arbor, MI, USA.

Portney, P.R. and J.P. Weyant (1999), “Discounting and intergenerational equity”, Resources for the Future, Washington DC, USA.

Powder River Basin (2004), Coalbed methane website, www.powderriverbasin.org/cbm/pressreleases.shtml.

Pruess, K. (2003), “Numerical simulation of leakage from a geologic disposal reservoir for CO2, with transitions between super- and sub-critical conditions”, paper presented at the TOUGHsymposium, 12-14 May, Berkeley, CA, USA.

REFERENCES 245

235-250 References 18/11/2004 17:30


Read, P. and J. Lermit, (2003), “Bio-energy with carbon storage (BECS): a sequential decision approach to the threat of abrupt climate challenge”, paper presented at the InternationalEnergy Workshop (IEW), 23-26 June, Laxenburg, Austria,www.iiasa.ac.at/Research/ECS/IEW2003/Abstracts/2003A_read.pdf.

Ree R. van (2000), “Technical and economic data biomass-based energy conversion systems for the production of gaseous and/or liquid energy carriers”, report no. GAVE-9915, NetherlandsAgency for Energy and the Environment (NOVEM), Utrecht, the Netherlands.

Reiner, D.M. (2004), “Public perceptions”, presentation at CAN Europe meeting, 27 May, Brussels, www.climnet.org/CTAP/workshop2004/PresentationsWS2.htm.

Riahi, K., E. Rubin and L. Schrattenholzer (2003), “Prospects for carbon capture and sequestration technologies assuming their technological learning”, Proceedings of the 6thInternational Conference on Greenhouse Gas Control Technologies, 1-4 October 2002, Kyoto,Japan. Oxford, UK, Elsevier Science (Pergamon), pp. 1095-1100.

Rigg, A.J. (2001), “GEODISC: the search for geological sequestration sites in Australia”, paper presented at the 1st Annual Conference on Carbon Sequestration, 14-17 May, Washington,DC, USA, www.netl.doe.gov/publications/proceedings/01/carbon_seq/4a5.pdf.

RITE (2003), “CO2 geological sequestration project”, www.rite.or.jp/English/lab/geological/geological.html.

Roekke, N. (2003), “CO2 separation technology and gas turbine applications – key challenges”, paper presented at the Global Watch seminar, 29 August, Trondheim, Norway.

Rothwell, G. and B. van der Zwaan (2003), “Are light-water reactor energy systems sustainable?”, Journal of Energy and Development, Vol. 29, No. 1, pp. 65-79.

Rubin, E. and A.B. Rao (2003), “Uncertainties in CO2 capture and sequestration costs”, Proceedings of the 6th International Conference on Greenhouse Gas Control Technologies, 1-4 October 2002, Kyoto, Japan. Oxford, UK, Elsevier Science (Pergamon), pp. 1119-1124.

Rubin, E., A.B. Rao and C. Chen (2004), “Comparative assessments of fossil fuel plants with CO2

capture and storage”, paper presented at the 7th International Conference on GreenhouseGas Control Technologies, 5-9 September, Vancouver, Canada.

Rutkowski, M. and R.I. Schoff (2003), “Pre-investment of IGCC for CO2 capture with the potentialfor hydrogen co-production”, paper presented at the gasification technologies conference,12-15 October, San Francisco, CA, USA, www.gasification.org.

Sakhalin Energy (2004), ”Sakhalin II project. LNG or DME – the facts”, www.sakhalinenergy.com/project/prj_overview_lngdme_report.asp.

Salvador, C., D. Lu, E.J. Anthony and J.C. Abanades, J.C. (2003), “Enhancement of CaO for CO2

capture in an FBC environment”, Chemical Engineering Journal, Vol. 96, No. 3, pp. 187-195.

Saripalli, P. and P. McGrail (2002), “Semi-analytical approaches to modeling deep well injection of CO2 for geological sequestration”, Energy Conversion and Management, Vol. 43, No. 2,pp. 185-198.

235-250 References 18/11/2004 17:30

Sarv, H. and W. Downs (2002), “CO2 capture and sequestration using a novel limestone lagoonscrubber”, paper presented at the 2nd Ohio carbon dioxide reduction, capture andsequestration forum, 26 April, Lancaster, OH, USA.

Scott, D.H. (2001), “Improving the competitiveness of next generation coal-fired plants”, report no. CCC/52, London, UK, IEA Coal Research.

SFA Pacific (2002), “Summary of bases and results of spreadsheet analyses of new coal based PC and CGCC power plants”, www.sfapacific.com/.

Shaw, J. and S. Bachu, (2002), “Screening, evaluation and ranking of oil reservoirs suitable for CO2-flood EOR and carbon dioxide sequestration”, J. Cdn. Pet. Tech. 41, No. 9, pp. 51-61.

Simmonds, M., P. Hurst, M.B. Wilkinson, C. Watt and C.A. Roberts, (2003), “A study of very largescale post combustion CO2 capture at a refinery & petrochemical complex”,www.co2captureproject.org/news/documents/

Singh, D., E. Croiset, P.L. Douglas and M.A. Douglas (2003), “Techno-economic study of CO2

capture from an existing coal-fired power plant: MEA scrubbing vs. O2/CO2 recycle combustion”, Energy Conversion and Management, Vol. 44, No. 19, pp. 3073-3091.

Sinor (2004), “DME synthesis could unlock remote coal reserves”, Sinor Synthetic Fuels Report, Vol. 7, no. 2, pp. 82-84, http://edj.net/sinor/sfr4-00a8.html.

Smith, A.R. and J. Klosek (2001), “A review of air separation technologies and their integration with energy conversion processes”, Fuel Processing Technology, Vol. 70, No. 2, pp. 115-134.

Stefansson, O. (1999), “Economic aspects of geothermal development”, paper presented at theinternational workshop on direct use of geothermal energy, 16-17 November, Ljublijana,Slovenia, www.geothermie.de/wa/Economic aspects of geothermal development.htm.

Stein, V. and T. Foster (2001), “Ceramic membranes for oxygen production in Vision 21 gasification systems”, paper presented at the gasification technologies 2001 conference, 7-10 October, San Francisco, CA, USA.

Stevens, S.H. (1998), “Carbon dioxide injection for enhanced coalbed methane recovery: San Juan basin test results and potential for China”,www.coalinfo.net.cn/coalbed/eng19984/423.htm.

Steynberg, A.P. and H.G. Nel (2004), “Clean coal conversion options using Fischer-Tropsch technology”, Fuel, Vol. 83, No. 6, pp. 765-770.

Swisher, J.N. (2002), “Cleaner energy, greener profits: fuel cells as cost-effective distributed energy resources”, Rocky Mountain Institute.

Takeuchi, M., N. Matsumuya and S. Niwa (1999), “Study on CO2 global recycling system”, Proceedings of the 4th International Conference on Greenhouse Gas Control Technologies,30 August–2 September 1998. Interlaken, Switzerland. Oxford, UK, Elsevier Science(Pergamon), pp. 433-443.

Thambimuthu, K. (2003), “CO2 storage and capture technology roadmap”, www.nrcan.gc.ca/es/etb/cetc/combustion/co2trm/pdfs/co2trm1_kthambimuthu_strawman.pdf.

REFERENCES 247

235-250 References 18/11/2004 17:30


Tomski, P. (2003), “Zero emissions technologies for fossil fuels. Technology status report”, appendix 1 : R&D visions. Preliminary draft update, June 2003,http://www.cslforum.org/documents/TSRAppendix2003.pdf.

Torvanger, A., S. Kallbekken, and K. Rypdal (2004), “Prerequisites for Geological Carbon Storage as a Climate Policy Option”, report 2004:04, Cicero, Oslo, Norway.

Tweede Kamer (2003), “Beleidsnotitie schoon fossiel” (Policy outline for clean fossil fuels), Tweede Kamer vergaderjaar 2003-2004, 28241, no. 6. Dutch Parliament proceedings, TheHague, the Netherlands.

UDI (2003), World electric powerplants database, Utility Data Institute. McGraw- Hill, Washington DC.

UN (2003), Industrial commodity production statistics. Dataset 1950-2000. United Nations, Statistics Division New York, USA.

Vendrig, M., J. Sponge, A. Bird, J. Daycock and O. Johnsen (2003), “Risk analysis of geological sequestration of carbon dioxide”, report no. COAL R246, DTI/Pub URN 03/1320,Department of Trade and Industry (DTI), London, UK.

Viswanathan, R. (2003), “Boiler materials for USC plant”. Electric Power Research Institute (EPRI), Palo Alto, CA, USA.

WADE (2003), “World survey of decentralized energy 2002/2003”, World Alliance for Decentralized Energy, Edinburgh, UK.

Wakamatsu (2004), “Brief overview of gasification and the status of our gasification project (EAGLE)”, Wakamatsu Research Institute,http://eneken.ieej.or.jp/japac/document/2110307170916_2.pdf.

WEC (2001), “19th survey of world energy resources”, World Energy Council, London, UK,http://www.worldenergy.org/wec-geis/publications/default/launches/ser/ser.asp.

Wilkinson, M.B., J.C. Boden, R.S. Panesar and R.J. Allam (2001), “CO2 capture via oxyfuel firing: optimization of a retrofit design concept for a refinery power station boiler”, paper presentedat the 1st National Conference on Carbon Sequestration, 14-17 May, Washington, DC, USA,http://www.netl.doe.gov/publications/proceedings/01/carbon_seq/1b3.pdf.

Wilson, M. and M. Monea (eds) (2004), “IEA GHG Weyburn CO2 Monitoring & Storage Project. Summary Report 2000-2004”, Petroleum Technology Research Centre, Regina, Canada.

WWF (2004), “Position paper. Carbon capture and storage from fossil fuels”, http://www.ngo.grida.no/wwfneap/Publication/Submissions/OSPAR2004/WWF_OSPAR04_CO2_Annex.pdf.

Yamashita, K. and L. Barreto (2003), “Integrated energy systems for the 21st century: coal gasification for co-producing hydrogen, electricity and liquid fuels”, report no. IR-03-039,International Institute for Applied Systems Analysis, Laxenburg, Austria,.

Zhou, Z., Y. Zhang, J.W. Tierney and I. Wender (2003), “Hybrid zirconia catalysts for conversion ofFischer-Tropsch waxy products to transportation fuels”, Fuel Processing Technology, Vol. 83,No. 1/3, pp. 67-80.

235-250 References 18/11/2004 17:30

Zwaan, B. van der and A. Rabl (2004), “The learning potential of photovoltaics: implications for energy policy”, Energy Policy, Vol. 32, No. 13, pp. 1545-1554.

Zweigel, P. and E. Lindeberg (2003), “Leakage rate – the main quality criterion for underground CO2 storage”, paper presented at the 3rd Nordic mini-symposium on carbon dioxide captureand storage, 3 October, Trondheim, Norway.

REFERENCES 249

235-250 References 18/11/2004 17:30

235-250 References 18/11/2004 17:30

You can also sendyour order

to your nearestOECD sales point

or throughthe OECD online

services:www.oecd.org/

bookshop

CUSTOMERS INNORTH AMERICA

Extenza-Turpin Distribution56 Industrial Park DrivePembroke,MA 02359, USAToll free: +1 (800) 456 6323Fax: +1 (781) 829 [email protected]

www.extenza-turpin.com

International Energy Agency9, rue de la Fédération75739 Paris Cedex 15, France

All IEA publications can be boughtonline on the IEA Web site:

You can also obtain PDFs ofall IEA books at 20% discount.

Books published in 2002 and before- with the exception of the statistics publications -

can be downloaded in PDF, free of charge,on the IEA website.

www.iea.org/books

The Online BookshopInternational Energy Agency

CUSTOMERS INTHE REST OF THE WORLD

Extenza-TurpinStratton Business Park,

Pegasus Drive, Biggleswade,Bedfordshire SG18 8QB, UKTel.: +44 (0) 1767 604960Fax: +44 (0) 1767 601640

[email protected]

www.extenza-turpin.com

IEA BOOKSTel: +33 (0)1 40 57 66 90Fax: +33 (0)1 40 57 67 75

E-mail: [email protected]

251-252 18/11/2004 11:46

IEA PUBLICATIONS

9, rue de la Fédération 75739 Paris Cedex

Printed in France by ACTIS(61-2004-291 P1) ISBN 92-64-108-831 2004

251-252 18/11/2004 11:46